'1'1' 1'11 " 31;;1. 3.I'. 'I!!"|' " '1; ' l .1 ,;33 "1;" ."'1'.. I 1111'" "11"; ,I'H'1 11' '1 ' I; 1111111111 1 1 1111131111211" ""1I(I11;| "311%? 73'13'1111/11/ 1 1 1111313113?” I 113; ; “1111111;334113111’1111'31113'; I1131/ 1101' 113113111: 1;;111111111311'131'3131 11111 '1' I"."I'I%, II' "|""" "'..-'3."1".',5I "11II;1'1133;.'I';1.115'1'3'3'1H111'3'31'11'113111'1‘1'3"; 1;11;JI1'1131.1I;I1H;I 1.”; ; _'.1'""'31"' 1'1' 1"11" "'I'|3""I"31171011.». 'I1.1 '311'3'3"1' 1." '3‘ 331311|:'31'31"3131313l331'."'1|"|"\'"1']'1" 1.1131111 3111'1'3'311'; '! J.1311'1'1'111.3111". I"';'1'|"1""/ ""'"|' ""1'I'1'1 '.‘ ' I. "'II." I... .I'I 13/ I'111113I 11111;:333333 1; '13131331'1' 1'111;11"‘"" I I31; 1- 111131331; 3.13."; “1 11" "1'" 1'1111'II"; -;; '1 11/11" II'3111;I'. "3" 1'1"! I'3I'I;;13 3113111111; 31113. '1"' "" "."""1';"I"‘.I'Im 3 ""3111" ,I' "" " 111,111; '1 , 33 3 3 13 '.'1'1' 1111.! rip! 11; "1111’: " 11111.1; —‘_ B__ “— ‘2; ‘2‘ 1 .. 0‘ g - 3c _‘ A \_ —': —-' k '- 2. -_4. 7:73: ‘2‘ w - 2' ‘.‘ - '11:; . 11111 J "1' "'1'"' 'J""' ' 1311"" I I | .. ' " I' 33.13. ‘1'11'...'311;3|3I33,; 1'1'1'11"'3;1"1"'I"1'13"11 1 1; 1 '13 I.'I,.I13,'. . ""11""ll "!|"" '1131'1311'H1' "' '1'”! '1 I}; ;;" "')3','" 31'1'3I31'I3131311 '3'11‘3, 3 "1".1 .""|I"'I' .‘ .1 '11,. 'I. "I. 113'3' ;. 33.33.133.313 1-3 33 I;'I3 .'I 'II. l3113!.I1II '1. 1131713131.. .3131. 1 ' 131131"3'3I3. I"1""'"'11'1. 3 .1; I'.'!iI/1';'1I;'.; 3' 3131131 ,‘ 3'13." 3 .' 3 3| 11."; 3 1 I " "' ""' """/"" "' ' "' ' ""1J';'1.'11;.""" '1;§'"1'33'33"'.' "7" 1'" 1" 3 1. 3.. .I' -'." ;I".'IIV3 ' I' '3'I"1.‘3|" I".|'| '1111111 11h 3 11";1 '3'1'3I1'I'I' "' .3 1"] I' '.'. "' .31: ' I... I». ,, .1 :II , I; , . I I.I.I~I|I .1. I', ..3.333;; ., I:.1'l31 '11117" |""'1'3' 3' 1 1 . 1 ||"31" 11.11, 13" 1'11; I" "1 ‘ 11""'I"'"I|'I1"'I .I, 1;"; I1"' "'"1""v' ' .‘r'. ' "'""' '.'"11'" I ‘ ' ' " 13'3; {13"1'11 3.1‘1I13 I II I 11I11'3" '31))'11I3IJ 11,! lr'ld' ('II, 1 ' ' 11" "I '1?" [11' "'1';"I""(1v‘ ' '11.. 1' i 1 11 1 1 3 3’3 1"""3\""'3I.'1 313313-1313 "333 '.31'11'1'31'33‘ 1“ 1 |_13:’;"3'313'3‘331~:3 ‘1 13 13:13 1". u ' ' """"""""""" .. ".'|-" ‘-"""" 'I '1." "' .' .I'I'"I. ' i;",I"':'I 1.31111" L {111111'1' .345}; {"1 @9111. " ' 'I' ..I ‘.'III' .31.; 'II... 11111,)...1 '1'; 13'3 ‘ 1‘ '3I"'3.I 3"I'I'31.'.'.'. .' 3'; ;‘,. 3 t3 ' J “111.13." 3| ;" . ! I 'l ' ' " ‘ " ' " ' ”I """TIIII' 11%;.” ”WM" """11' "" 111;"1'3 '. . "'. 'I"".' I .1 1 " " 1'1 1 11"II ""1'I‘1I!1"""“ ‘ "" ' "11'1“ 1' I"-. I. II ‘ ‘ '3' ..'| .3 ; ' 3.31.1.1 ' "". "1":1 1.5 '731JI13111L'1I! M1" . . . 1' . . . . 3. . . .. 1114'. ' .'I'“. I . ‘ .'.'. "' '. '.' "I ‘ 1'1" """'I""I" . ' "'.'” .I'I 1.; ..."1‘ 5111131111"? 1111' 1111 "um" ' '1' 1. 1 I "1'" III-I ' I . 3l1l" 1". I.'I.'I"' '1" ' ...33I1.I""!" "" ,3" "I .' I. I." ".' '. ". . ' " “1";1I'.’ 'II' II, "'1'" 1" " I"1'.2‘..'." ""1 "1:1,. ' '. 1' I I! '1"; .'I "" " 1" 1'1';' """I11'1" "" ‘13 '13‘1"I‘ " 'fl"""“"',"'.l f". ')‘ ‘ ""'.I""1'| I J ‘1'1-1'1 '.'.3' I" 3, 3 . 3 3 '3 l 1 I 1" ' 31J ' .13'_ 3 ! . v11. .' 3n", ~ul' ‘ “v-1 :5“ I: 1 ' ' "J" " ', | ‘ | . .1;' I" "".| ' """11" ,.. .1'.I'V 1'3‘3.;."1.'.,,..-' . |~ " " .. :'.,'-.'..‘. I'I"‘-'.I1:j'£"1.u1I'f'lw'. ‘3"'!;' 3"' 3' ' 3 H 3 3 3 ';' II ' " “31'1”. ' ".331 ' 3,1' 1' .3133 III . . .'. '3'!‘ “It.” 'yll' ,f'1'II‘L" 3":(fl“3\’\'\l-’1 “WM" II" ' ' ' 33' 3" 1 .' I'.' . 'IJ'I’JH' 'I ; I" '<' 3.'II 3"'1'1'1' I . ' " ' ' 3 ., 1 "J '3 '31'1 ‘7'111'...‘0“. '111 "1'1"!“ 5 ' ‘ " 1 ' ‘ ' '1 'III. III " I '1" .'"I .I" 1 . 11' . . I- I. ‘II' "'2 ' '!'-'I‘ -' 1" .'. I " ’ I ' ' ' ' " "I ." " """| ""'1"' I' "I !""'1""""" I'1'II1' 3" ¥LO ' 1'" 33 ,1 . 1, 1 . “ , 1’. " "I'l‘! 3 .. '.' I 33' 3‘ ' 3.,1. ',. ;'I"3I'I3 .' 3113! 1131'33,‘.. ' 3 33. .'3 .11 1. 3 3 | 1 1 . l 1' '3 33 111‘ ‘1u1 ‘II J '3“ ""' " '.' ' .I‘III' .I I. " ' . 11:! "." ;;.. 'I" " “'."IAJIV " 1:! '13 '3 'l;, . 33 '13II131' . 33'3' 3 I 1 ' ' 3 ‘I3I313‘ 1.1 I "3333 11'l11' “.11'_ ."I " '1 'I I I ' I " ' l ' 1" #11 ' ‘|‘ ! ' ..3I'.‘ I ‘ '1‘ , . . ' ' 1; “‘ ‘ ' 3 1f“, 'II""1) 1 " 33.33133 ' 3 33 ...3 .1 l ‘3 31 1 l I . ’. ' 1 !‘ I!I!3!|I ‘ . 3 31\ ?|. 3. . I‘ 3 |‘ I.. >.,' ' 11.; I 3 I Il‘IIQI‘ 1““ 'I'|'|" ' | . . K """'~'|I""1'.' "‘ ' ' ""|- . '. | . ' I I" ' I ' !I'II"""""I'I'1""' '1' I ' ' I3 . . 3 3 I . .1 11 '1'3l'1"' ' '1 I, II3I. 3; I |.|IIII n' "'1. "'mu'h IIIIII-IllldIIfnL1' ' "1"" 'I' 11 I I1'1' " ' ' ' 111' '1 H ' I . I1"1' T.‘{ «.'Q 9r""‘ 4 #“J‘ "9 q I ' LIBRARY Michigan State University This is to certify that the thesis entitled SHIP TRAFFIC SIMULATION AND PORT INVESTMENT OPTIMIZATION FOR LAGOS PORT COMPLEX IN NIGERIA presented by Samuel Kingsley Nnama has been accepted towards fulfillment of the requirements for Ph. D. Civil Engineering degree in .——-’ Mézgm {z , /,;ug 4‘». Major professor Date July 31, 1979 07639 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records -* we? M . m 113-: :w‘ - l” ., “:4 ‘Q‘M’é’m-g' L 2‘. ‘. k'# ‘lir‘fcvkzj 5 " . . ,- “-1.". ,Lffi ® Copyright by SAMUEL KINGSLEY NNAMA 1979 SHIP TRAFFIC SIMULATION AND PORT INVESTMENT OPTIMIZATION FOR LAGOS PORT COMPLEX IN NIGERIA By Samuel Kingsley Nnama A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Civil and Sanitary Engineering ABSTRACT SHIP TRAFFIC SIMULATION AND PORT INVESTMENT OPTIMIZATION FOR LAGOS PORT COMPLEX IN NIGERIA By Samuel Kingsley Nnama The objectives of this dissertation are three fold; first to ana- lyze and simulate the ship queuing problem at Lagos port. Secondly the research will prepare an annual cost analysis of various logistical port subsystems. Thirdly an optimization program will be employed to establish a sound criteria for port investments. To achieve the first objective, the Chi-square statistical technique was employed ‘to test the ship arrival and service dis- tribution. Ship arrivals at the port follow a Poisson distribution while service fits a negative exponential distribution. A Fortran program was written to simulate the ship queuing process and print ship delay parameters and queue length. The optimization program was employed to determine the optimum combination of port resources considering total thoroughput and total cost to the port under specific time constraints. In general.this dissertation focuses on the problem of port con- gestion in developing countries. The overall approach is to analyze the port problems by relating it to national economic growth and other multi-modal transportation systems. Hence the author carried out a detailed discussion of multi-modal transportation networks in Nigeria. Samuel Kingsley Nnama However the methodology adopted in this research can be applied to any general cargo port both in developed and developing countries. As each general cargo port has unique morphology and service policy care should be taken in identification of traffic variables and logistical subsystems. DEDICATED TO THE MEMORY OF MY FATHER AND MOTHER CHIEF JOSHUA ORJI NNAMA THE OFLOZOR IV OF NIBO, NIGERIA AND MRS. SELINA MGBAFOR NNAMA THE ORIMILI I OF NIBO, NIGERIA In recognition of your singular courage to break with established palace tradition in 1919. You chose to become missionaries. In four decades you established numerous schools and churches in several towns in both the upper Niger provinces and the impregnable Delta provinces. By doing this you unveiled the blanket of darkness and brought light to millions of less privileged people. It was in recognition of these services that you were nominated as a life member of the Niger Anglican Synod and the Niger Mother's Union respectively. Your example of princely humility has been unprecedented in the Niger Diocesse. On behalf of your six children may I eXpress our gratitude for the great price you paid for our education and up bringing. We have always counted your benevolence, honesty, steadfastness and calm composure as our greatest heritage. ACKNOWLEDGEMENTS I wish to express my gratitude to all the members of my Ph.D. guidance committee. Dr. William C. Taylor has been particularly helpful from the initial process of proposal devel0pment to the completion of this research. He shared the challenging problems of this dissertation with the author. His ideas provided direction when all hopes were lost. Dr. James Brogan in addition to serving on my guidance committee was my major supervisor as a graduate research assistant. The projects provided immense training for me. Dr. James Brogan also followed this research with special interest. I am grateful to him for useful ideas in developing the research structure. Dr. George Wagenheim has been my major advisor in areas of system logistics. He was instrumental in developing an operations research orientation to this problem. Dr. Gail Blomquist was readily available to make contributions on research problems. His advice helped me to determine the scope of the study. I am grateful to the entire members of my Ph.D. guidance committee for their numerous contributions which cannot be mentioned. My gratitude also go to Chief and Mrs. Godwin Nnama, the Managing Director of NNAMA Shipping Lines, Inc. (NTL), for providing me with a scholarship up to graduation from the University of Nigeria Engineering School. The summer job experience I earned in the NTL introduced me to the logistics of ship traffic and cargo flow. My twin brother, Emmanuel and his wife Mercy deserve my special gratitude for their encouragement iv and support during the difficult days of this research. Emmanuel as the Assistant Chief Port engineer for Lagos port was instrumental for quick data collection and Operations survey forms circulation. I will also use this opportunity to thank my sister Dinah and my brother Joshua for their contributions to my education. The Rev. Beford Nnama, my immediate senior brother deserves a special gratitude for his numer- ous contributions and encouragement at various stages of my education. My wife Ihuoma has shown unlimited devotion and understanding during the field studies and the entire period of this research. She has been a tower of strength during my period of grief. I share this accomplishment equally with her. The birth of my son Ifeanyi Sam Jr. generated a new vigor in me therefore he equally deserves a special mention. I owe unlimited gratitude to Ihuoma and Sam Jr. I wish to thank the General Manager of the Nigerian Port Authority, Alhaji Bamanga M. Tukur for his cooperation and authorization of this study. I am also grateful to numerous officials of the Nigerian Port Authority who were involved in various aspects of data collection and operations survey. Finally I wish to thank Brigadier Benjamin Adekunle (former military commandant of the port of Lagos and former commander of Nigeria Elite Commando Division) for useful discussions on the past and present problems facing the port of Lagos. Brigadier Benjamin Adelkunle lead a successful military campaign to decongest the port of Lagos in 1970. He is reputed as one of Africa's leading military logisticians and tacticians. His interest in this research has been Special and I wish to thank him for this. List of List of List of CHAPTER 2 —i i—H—H—H—H—I Hm th—Ig 0145me 35 00 out! NNN'U o o H o o o N '50 CHAPTER f’ H O I > UlU‘lU'I'U h-b 99-h ._. town-am OlU'l th x CHAPTER 6.1 6.2 6.3 6.4 CHAPTER TABLE OF CONTENTS Tables Figures Appendices I -- THE PROBLEM Introduction Overview of Nigerian Economy Ocean Ports in Nigeria The Existing Problem and Scope of the Study Economic Justification for the Study II -- NIGERIAN ECONOMY Nigerian Economy and Population Growth Trend Multi-Modal Transportation Systems Demand for Shipping Services in Nigeria III -- LITERATURE REVIEW Literature Review of Current Trends in Port Planning and Analysis Identification of Limits in Existing Models and Justification of the Dissertation IV -- FIELD STUDIES PROCEDURE Introduction Design of the Data Collection Froms Port Morphology Computation of Annual Investment for Various Logistical Subsystems Ship Arrival and Service Distribution Analysis of Cargo Delay V -- SHIP QUEUING SIMULATION MODEL Assumptions Logic Diagram and Model Variables Sensitivity of Total Ship Delay to Increase in Number of Berths VI -- INVESTMENT OPTIMIZATION The Objective Function Determination of Optimization Constraints The Optimization Process Traffic and Logistical Operations Survey VII -- ALTERNATIVES AND RECOMMENDATIONS APPENDICES BIBLIOGRAPHY vi Page vii ix \IU'I-bwv—I 34 55 75 85 103 104 105 108 121 136 146 147 152 159 161 172 198 214 225 257 Table H I g...- -rrrrrrr @WNOthNH 10 11 2-12A NN ll 2-128 2-13 2-14 2-15 2-16 2-19 LIST OF TABLES Ocean Ports of Nigeria Gross Domestic Product Second National Development Plan Third National Development Plan Comparison of Projected Food Supply and Demand Comparison of Gross Domestic Product Exports of Principal Commodities Agricultural Exports National Budget Allocation 1978 Population Estimates Population Growth Estimates Population Densities by State Total Investment in Transport Sector by All Governments 1970-1974 Nigerian Railway Performance Safe Draughts During One Navigation Season on the Benue River Benue Riverports Freight Rates for Selected Commodities for Different Transportation Modes Internal Airports Ships Entering Ports by Produce Seasons Seasonal Variation of Nigeria's Export and Import Trade Average Seasonal Import and Export Trade Possible Benefits of Port Investment M/M/C Steady State Equations UNCTAD Definition of System Parts for Port of Casablanca Lagos Port Complex Annual Investment Cost Breakdown: Transit Sheds Annual Investment Cost Breakdown: Warehouses Annual Investment Cost Breakdown: Dredging Annual Investment Cost Breakdown: Paved Storage Space Annual Investment Cost Breakdown: Berths Annual Investment Cost Breakdown: Cargo Handling and Equipment Annual Investment Cost Breakdown: Tugs and Barges Arrival Distribution Chi Square Test Applied to Arrival Data Ship Service Distribution Chi Square Test Applied to Ship Service Time at Berths Cargo Delay (Parameters for Time Functions) vii 109 110 111 112 113 114 119 120 124 129 130 135 139 144 Table 01 Oh mmmmmmmmmm @0101 03 01 \IN NH Sensitivity of Delay to Increment in Number of Berths Sensitivity of Average Ship Service Time to Increase in Number of Equipment and Labor Gangs Optimization Constraints Summary of Optimization Constraints Sensitivity of Average Ship Service Time to Increase in Number of Equipment and Labor Gang (Work Study Data) Summary of Optimization Results for Case 1 Summary of Optimization Results for Case 2 Summary of Optimization Results for Case 3 Cost Effectiveness of 1978 (Short Term) Alternatives Summary of Optimization Results for Case 4 Summary of Optimization Results for Case 5 Summary of Optimization Results for Case 6 Cost Effectiveness of 1990 (Long Term) Alternatives Response to Logistical Operations Survey Forms Analysis of Traffic and Logistical Operations Survey Forms (Category A: Traffic Officers) Analysis of Traffic and Logistical Operations Survey Forms (Category B: Port Operations Officers) Analysis of Traffic and Logistical Operations Survey Forms (Category C: Shippers) Table of Distances Port to Port (Nautical Miles) Cost Effectiveness of Alternatives viii o—IHkOOONOSU'IAw NH HO I NHt-l HNH 010101 hhh 000000 NNNNNNNNN NM 0 o o o o o o o o o o o o o o o o o o 0 (”NM 01010) “NH 0‘ LIST OF FIGURES Location of Nigeria's Sea Ports Map of Africa Economic Resources of Nigeria Percentage Allocation Under Recurrent Budget 1978 Total Investment in Transportation by Federal and State Governments of Nigeria 1974 Transportation Map of Nigeria Rail Network in Nigeria River Transport Map of Nigeria Nigerian Airways Internal and External Flight Routes Ship Traffic Entering All Nigerian Ports Ship Traffic Through the Port of Lagos Lagos Ship Demand Against National Demand Demographic Map of Nigeria Road Map of Nigeria Poisson and Erland Distribution Curves Port Control Volume Configuragion of Ship Queuing System Lagos Port Location Map Cumulative Ship Arrival Distribution Cumulative Ship Service Distribution Simulation Logic Diagram Lagos Port Simulation Model Sensitivity of Queue Waiting to Increase in Number of Berths Sensitivity of Berthing Time to Increase in Equipment and Labor Cost Sensitivity of Que Waiting Time to Reduction in Berth Service Time (1978 Demand) Sensitivity of Que Waiting Time to Reduction in Berth Service Time (1990 Demand) The Proposed Rail Network Linking the Port of Warri and Koko to the National System ix Page 176 179 190 219 LIST OF APPENDICES Page Research Data Collection Forms 225 Computer Output of the Lagos Port Simulation Model Developed in this Research 242 Synopsis of the Port Simulation Model Designed by United Nations Conference for Trade and Development (UNCTAD) 249 Background of the Author 255 CHAPTER I 1.1 Introduction The international demand for shipping services is based on the economic law of supply and demand. Developing countries export agricultural products, timber, minerals and crude oil to the in- dustrialized countries in return for processed food, drugs, manu- factured goods, equipment and military hardware. This trend in international commerce creates a two directional logistical flow. Ocean shipping provides the dominant mode for movement of goods be- tween nations. Seventy-eight percent of the total tonnage in the- world trade are moved by merchant marines. Airlines move .5% of the total tonnage. The balance of 21.5% are moved by overland carriers rail and truck lines between contiguous countries. (1)‘ Ocean shipping is dominant because of the following major compara- tive advantages: Ocean liners and tankers have tremendous freight capacity (20,000 - 60,000 tons). A large liner of 60,000 tons moves six times the freight accommodated by a train load of 66 cars. Shipping freight rates are very low. The average revenue per ton mile for all modes in international trade is shown below Ocean liners - .2¢ per ton mile. Rail - 1.3¢ per ton mile. Air - 29.9¢ per ton mile. (2) 2 The tendeney is for shippers to move only costly inventory (e.g. jewelry, watches) by air. Since the rail lines are limited by the geographical dispersal of the continents, ocean liners are expected to remain the dominant mode for providing international trade. This view is substantiated by the fact that both international trade and ocean shipping are growing at almost the same rate. Between 1950 and 1970 world trade value increased by an annual rate of 5.42% while annual waterborne tonnage increased by 4.67%. (3) It follows that the smooth flow of goods will require the development of well equipped national ports. This need is even greater in developing countries because of the dependent nature of their economy on imported goods. Developing Countries: Developing countries generate 41.0% of the world seaborne trade. (4) In recent times earnings from crude oil exports have bolstered national foreign exchange reserves. This has created.afavorable balance of payment position for some develop- ing countries. The result is that the level of import rises. This trend is expected because imports are a function of exports (i.e. input-output model). As those increases occur, port logistical sub- systems (e.g. berths, shorehandling equipment, warehouses) become inadequate to handle the increasing number and types of ships. A common result is port congestion and ship queueing. This implies that port planning and development should be tied to the economic growth of developing countries. Planning of ports must be carried out as a long term measure and not just an ad hoc project. The objective of this research is to analyze the planning requirements at the port of Lagos. 3 1.2 Overview of Nigerian Economy An overview of Nigeria's economy is important to highlight the impact of economic growth on the port of Lagos. The Nigerian economy has been growing very rapidly since 1960 when she gained independence from British Administration. The discovery of petroleum onshore and offshore in 1957 was an economic landmark. Higher prices of oil in world market increased Nigeria's foreign exchange holdings. The gross domestic product jumped from $3.3 billion in 1960 to $4.8 billion in 1970. (5) In addition to the petroleum increases, agricultural and manu- facturing sectors have been growing at the average of 4% per year. The emphasis are on mechanized agriculture and large scale farming. Intermediate industries are being established for the processing of agricultural products and manufacturing of durable goods. Governmental service is the highest growth area, with an annual growth rate of 30%. (6) The Government policy is to provide better education and health services to the people. This entails construc- tion of numerous physical infrastructures to accommodate the expansion. These increased Government service activities have an upward multi- plier effect on the national economy. Employment is increased in all sectors and a boom is created in the construction industries. Thus with more public and private spending the quantity and quality of imports is increasing. In total, the Nigerian GNP has been growing at an annual rate of 10.0%. (7) This rate is significantly high for a developing country. It has many desirable features, but it is accompanied by a major problem, i.e. modification of physical infrastructures to 4 accommodate the congestion created by such growth. National ports are examples of such infrastructures which must be developed to service increasing numbers of ships andto handle higher import and export freight tonnage. 1.3 Ocean Ports in Nigeria Figure (1.1) illustrates the location of Nigerian seaports. The major seaports are Lagos and Port Harcourt, which together handle about 60% of the total freight tonnage. These two ports are general commodity ports and are equipped with specialized terminals for handl- ing refined bulk products. The port of Lagos is dominant because Lagos is both the comnercial center and administrative capital of Nigeria. This port is also a gateway to the densely populated Western States. Port Harcourt is a natural sheltered deepwater port. It services the oil rich delta region. Burutu, Calabar, Warri, Sapele, Koko and Degema are minor ports located on inland channels which require regular dredging. The fol- lowing table indicates the type and percentage of total freight tonnage (import and export) handled by each port. The port of Bonny is a specialized port developed to handle only crude oil shipments. The crude oil terminal is located to service ocean tankers directly. On the other hand, the minor ports of Sapele, Warri and Koko are inland river ports which are influenced by seasonal variation of channel depth. In low water (between December Ist and April 30th) the Benin and Warri rivers are 15 feet deep. (8) This means that only medium liners (20,000 tons) can safely navigate the channels. 5 Table (1-1). OCEAN PORTS OF NIGERIA (9) PORTS % of Total Tonnage No. of Berths Remarks Major Ports LAGOS 50 39 Major General Commodity Port PORT HARCOURT 10 12 General Commodity Minor Ports Sapele 1.5 3 General Commodity Warri . 2.0 3 General Commodity Burutu 2.0 3 General Commodity Calabar 2.0 3 General Commodity Degema 1.0 2 Palm Oil Port Bonny 30.0 6 Crude Oil Port Koko 1.5 3 General Commodity In conclusion, the ports of Lagos, Port Harcourt and Calabar are more suitable for major expansion because of year round deep approach channels (25 feet) and shipper preference because of their location. 1.4 The Existing Problem and Scope of the Study The port of Lagos was built by the British administration in the early twenties. Nigeria's exports prior to 1957 were mainly agricul- tural products (_c.ocoa,palm oil, groundnuts, rubber and timber, etc.). With the production of oil the national balance of payment position became more favorable. This economic boom resulted in greater govern- ment and private spending which required higher import levels. In 1964 the port of Lagos handled imports of .99 million ton; ten years later import tonnage through the port increased to 2.29 mil- lion. (10) In this time the physical facilities at the port did not expand much. The number of berths became inadequate for the in- creasing number of ships calling on the port of Lagos. An average of ten ships a day arrived at the port of Lagos despite the fact that all 14 berths were occupied 95% of the time. (11) 4 Ship service times at berth were very long; it took an average of 10 days to off load and reload a ship. The delay was caused by lack of modern shorehandling equipment (e.g. cranes, fork lifts, roll on roll off steel structures, elevator conveyor belts). Warehousing space was not adequate to provide for increased freight-tonnage. This situation required that vessels (except container ships) had to wait until ware- housing space was available. Harbor masters were unwilling to discharge cargo outside warehouses because of the high degree of pilferage asso- ciated with the port of Lagos. In the rainy tropical climate probabil- ity of damage is high to uncontainerized cargo. Since all port facilities in Nigeria are government owned and operated, there was no opportunity for private investments to increase the logistical subsystem within the port complex. Another bottleneck was the lack of efficient multi-modal freight terminals for the flow of goods in and out of the port system. Lagos port was designed as a railway port but train operations were irregular due to the lack of rail units allocated to port operations. The trucking industry dominated freight movement in and out of the port. This mode hauls 80% (12) of the total import freight leaving the port complex. It is apparent that this mode does not meet the high 7 capacity required by port operations. One thousand trucks are re- quired to move freight offloaded from an average liner of 20,000 tons. The overall results of these structural and technical inadequacies can be summarized as follows: (13) : Delay in ship waiting time within the queue. Average delays in 1975 was as high as 60 days. - High berth occupancy rate. All 14 berths were occupied 95% of the time in 1975. - Delay in ship service time at the berth. An average of 10 days was required to offload and onload a ship. - Ship cluster off the coast of Nigeria. In July 1975 ships in the queue numbered 230. - Due to high berth occupancy rates express ships were forced to wait for one or two days. Demurage on express ships are as high as $3500.00 a day. 1.5 Economic Justification for the Study In 1975 a cluster of 230 ships waited an average of 60 days before berthing. The Government was worried because of the economic impacts of this situation. The average cost of demurage for general cargo ships was $2500.00 with some ships' demurage as high as $3000.00. (14) The total annual dollar cost of demurage can be estimated as follows: Total Cost = Cd x T x N where Cd = Daily Average Demurage fee for one ship in the queue. T = number of days a ship stays in the queue before berthing. N = total number of ships which entered Lagos port in 1975. $2500 x 60 x 4000 $600,000,000 Total Demurage Cost 8 This figure represents 14.5% of the national foreign exchange holdings in 1975. It is evident that port demurage is draining the national reserve. The government, in an effort to capture some of the demurage charges, increased import and export duties. The ship- pers in turn increased unit cost of goods sold to consumers. This situation created inflationary pressure on the economy. To rescue the market situation, the government enacted price control laws. These laws were not very successful because many dealers resorted to black market sales. In-transit damages occur to some commodities due to long ship waiting times and rainy tropical weather. In 1974 four ship loads of cement valued at about $2 million spoiled while the carriers were in the queue for over 60 days. (15) The building industry suffered tremendously because of a lack of construction materials. Government construction projects (e.g. schools, hospitals, factories) were de- layed. Individual home builders had to abandon construction due to high costs of materials. In general port congestion created a down- ward multiplier effect on the economy. Unemployment and inflation increased due to the slump in the construction industry which is a major employment sector. Regional Influences: Port congestion in Nigeria has great regional influence on landlocked countries like Chad and the Niger Republic. These two countries direct 40% and 60%, respectively, of their export and import freight via Lagos. It is evident that these countries are also affected by congestion at the port of Lagos. The port of Lagos is a regional port and an important node center for the movement of import and export commodities. In July 1975 an agreement was negotiated between Nigeria and Ghana allowing ships bound for the port of Lagos to discharge their freight in Ghana's port of Accra. (16) Truck lines would complete the service by hauling the goods from Accra to Lagos a distance of 300 miles. This arrangement had specific problems; first Ghana de- manded exchange of crude oil in return for these services. Another agreement was signed with the Republic of Benin to enable overland shipment through its territory. Overland shipments did not work as efficiently as expected because of the tariffs associated with inter- national shipments and in-transit losses. Improvements In September 1977 the Nigerian Government completed the 'Tin Can' extension of the Lagos Port complex. This improvement provided ten extra berths with modern shore handling equipment and warehouses. (17) It is apparent that this improvement will reduce ship waiting and service time but the following questions remain unanswered: - What was the relative contribution of inadequate berths and poor service facilities at the old and new berths to ship delay and cost? - What is the optimum number of berths to meet the demand for shipping services in the next 20 years. - What logistical subsystems should be combined to achieve a flow rate that will minimize total ship delay? . What multi-modal systems are necessary for goods movement into and outside the port system? - What is the cost effectiveness of alternative port invest- ments (e.g. new berths, port handling equipment, multi-modal transfer facilities, rail road yard). 10 The 'Tin Can Creek' port extension was capital intensive, total cost were as high as $350 million. (18) However, this expansion was merely an ad hoc measure to meet the challenge of port congestion. The Nigerian Port Authority deserves credit for this improvement which averted a total failure. There is still a great need for fur- ther logistical and engineering analyses aimed at increasing the flow rate of goods and reducing ship delay. There is also a need to tie port development with the economic growth of the country. —4\‘ Vt. v \\4\ h h \\ 11 ooH mzzom zmo upsocoum 4o acumpcrz pmcmvmm .ommfiummmfi cop; acmEnoPm>mo Pacowumz newsh may sow mwcwpwuwzw ”mueaom nmm.fi . o~o.m~ mum.m~ oNc.H . o~o.oH mam.- m.~ mmo.mm Ppo comm :opmz m» NNH.~H . ouo.mm www.me om¢.m . Nom.mm www.mm N.o- o-.m~ pwo uzzuczogu Nu mo~.mom . Hm~.m~m nmm.~mm wmm.~ofll ome.wmm ewm.mmo N.o- cmm.mmm P_o spam #4 mom.mm . ume.mo~ omm.mm~ com.om . -¢.o- moo.~mfi m.m mmm.mm~ mpwzcm mm -o.m- n www.mmu.fi noo.smm.~ mmm.nofil m~o.~me.~ Hmm.mmm.H m.m nmo.¢mfi.~ mmpnmummm> Hm vm¢.o + mmm.~m ~mm.om m-.o + Ham.m¢ mx~.m¢ m.~ mmH.om mummm _::mm Nu mmo.o + m~o.o~ mem.mm umm.o + o~¢.~o mHH.Nc m.~ on.~m mummm cope: Ha Hmm.mm + www.mu mmm.Hm mNm.HH + mom.mm mmo.cv N.m e~m.mm mcmmm maom mu owm.mm~ + omm.mmm co~.m~m onm.m~fi+ mom.emo mmm.mom N.m mon.om¢ Ammazouv mcmmm No ¢-.~o~ . m-.~e~ Hom.omm mmw.~m . moo.em~ mmm.-m m.o- mm~.mm~ Hzcncaocw Ho “Nm.mm~ + om~.omo.~ mmm.m~m.~ omm.mm + mmm.mm¢.H NNH.moe.H m.~ mmm.omm.~ cpmgcmpa ma emm.mm + mmm.m~o.fi mm¢.mmm www.mm + mmm.mmm mmo.oom m.~ mm~.~om msm> ouou ma omw.~m~.fil mm~.~co.m mmm.mmw.m Hom.mmmu www.mma.m mm~.-m N.o- mmo.mmm.~ msm> an Hmm.Hmm + mom.m-.o~ mmm.mmfi.m mmm.~om+ mom.o¢m.m o~m.mm¢.m m.m mmm.~mm.~ e>mmmmu an om~.- . m-.¢m mum.Ho ~m~.~m . mum.¢~ mom.we «.0 emo.- poms: mm moo.mmm + ~m~.¢mo.~ o-.mmm 5mm.HHN+ mcm.nmm flow.mme ¢.oH vom.mmm mu_m em o-.wmv.fil mmm.m~m.~ wom.mom.v oom.nm~- ~mm.-m.~ Nme.o-.m m.o- oo~.mmm.m augmeom mm umo.m~m . “mm.m~o.~ ema.¢mm.m uoH.omN- meo.mum.~ ~m~.m-.m m.o m-.mom.fl poppy: mm www.cm . mmn.moH.H ~o¢.oo~.fl mfim.dm . -~.Nmm omo.emo.fi ¢.N mmm.~mw m~pmz Ha mo swzoga we uwuvmmo aPnaam uwuwemo apngam mama teach mmvpansm co umuumnoca newswo co umuumwoca ucoswo Pozcc< smm> mmmm mspaeam omma umuumnoee mapagzm mmoa emuumnoea uczoasoo mmumoma xgwuoesoo Aooo.v .oz4a¢=m coon ompomwozm no zom~m gm: ~w>pu cmwgmmwza .mmmH xoom me> paupampumum copemmvz “mocaom m.m 3.3 3.3 e.m m.~ 3.m “.3 3.3 m.~ 3.~ o.~ 3.3 3.3 o.~ moo3>eam eoeoo .33 3.o k.o e.o 3.o m.o 3.o w.o m.o ~.o 3.0 m.o e.o e.o m.o e33ao= .33 3.3 3.3 3.3 N.m 3.3 0.3 3.3 0.3 3.3 3.3 3.3 o.m w.~ e.~ ee33ao=e3 .o3 a.~ 3.3 3.3 3.3 m.~ m.m m.m 3.3 3.m o.m m.~ o.m 3.3 3.3 3eoEeeo>ec 3eeoeou .m e.c e.o o.o e.o e.o 3.0 mo. mo. e.o 3.0 m.e m.o e.o a.o ee33ao3e=eeau .m 3.3 0.3 3.3 3.3 e.m 3.3 m.m o.e 3.3 m.e 3.3 3.3 3.4 3.3 eteameeee .3 3.o3 3.33 m.33 3.33 m.N3 3.m3 3.33 m.N3 3.33 m.m3 m.~3 0.33 e.N3 3.33 ee33ee3e333e .e o.m 3.3 3.3 o.m 3.3 e.e 3.3 3.3 N.m m.e 3.3 3.3 3.3 c.e eewwwemwwmwwam .m m.o m.o m.o m.o e.o 3.o e.o 3.o c.o m.o m.o e.o e.o m.o ace 333onqu33 .e e.~3 3.33 m.o3 N.o3 3.3 3.3 e.3 3.3 o.k 3.e o.e e.m 3.3 m.e ee3e=3oae=eez .m v.33 m.O3 3.3 o.e 3.3 m.m e.e m.e m.e 3.3 3.3 3.~ 3.3 3.3 333° .3oe33 me3e32 .N ~.ee 3.ae 0.33 3.33 3.33 e.~m m.Nm o.~m e.mm 3.mm m.3e 3.30 ~.Ne 3.3a oee33=o3em< .3 33 33 NA ,3“ IO3 «me two 3e me me em me we 3e \33 \N3 \33 \03 \ae \ee \3@ \ee \me \ee \33 \30 \3e \0833 303833 333 3mee ee3oea mem3-~mm3 3< euaeoma U33mmzoo mmome do zo333moazou .Am-Nv o3aa3 23 - Poor feeder transportation systems required to connect remote farms to the market areas with high population concentration. - Land tenure system--particularly in southern Nigeria. This system involves ownership of small parcels of land by individual farmers which hinders the benefits of large scale farming. - Over 10% of the arable land require irrigation systems which are capital intensive. - Crop diseases and pests create major problems in the semi-arid areas of the country. - Federal and State spending in agriculture have not been sufficient to transform additional labor intensive farms to mechanized high output farms. The government acknowledges the existence of these problems; as evidenced by the policy on agriculture as expressed in the third national plan: "The conclusion to be drawn is that at the present rate of growth of supplies Nigeria will not be able to feed it's people in the next decade unless there is a radical departure from existing attitudes to investment in Agriculture." As a response to the above observation in the 1978 natiOnal budget a total of-N-2.2 billion was allocated to agriculture identi- fying it as the main single activity. A total of-N-39.00 million was allocated for the development of irrigation networks in the following river basins: Sokoto-Rima, Chad Basin and Funtuabagricul- tural project. Twenty four thousand tonnes of wheat, 30,000 tons of rice and 3000 tonnes of cotton (8) will be produced in these areas respectively. Four other River development authorities have been established. Mechanized farming cannot be achieved by government spending alone. The private sector has a great part to play. Hence there 24 is a need for Nigerian farmers to form cooperatives to enjoy economies of scale due to large scale farming. The Agricultural development bank should modify its existing tight loan credit qualification terms. This will make it easier for middle income farmers to obtain funds for improving their farms. There is also a need to develop feeder trans- portation systems to provide access to remote farms by linking them with urban markets. This will induce flow from areas of surplus to areas of scarcity. Individual farmers should be given incentives through reasonable prices. The present system whereby Government owned Marketing boards set prices without adequate farmer participa- tion provides no motivation to cultivate. This is one reason why 34 million hectares out of 64 (9) million hectares of cultivatable land are farmed. If more hectares of land are brought under cultiva- tion, the present food shortages should never occur. In 1970 food imports were valued at N-19 million. (10) This trend will continue to be upward unless mechanization of agriculture is carried out extensively. Extreme restriction of the import of food items is not feasible unless domestic production can keep up with grow- ing demand. Food import restrictions will only generate inflationary trend on the economy. Because of all the infrastructure and policy changes required to increase agricultural production, port planning should continue to be based on a rising or at least stable volume of imports. 2.1 (b) (ii) Oil Sector: The production of oil in Nigeria transformed Nigerian economy from a purely agrarian base to an energy exporting 25 economy. Nigeria joined the powerful OPEC organization, and in 1971 became the ninth largest world oil producer with a daily output of 1.68 (11) million barrels. In 1973 production increased to 2.3 million (12) bpd. (i.e. close to Libya which was the highest producer in Africa). By 1974 Nigerian daily oil production was 2.4 million (13) bpd. of high priced low sulphur crude. This figure represented 6.7% of the total OPEC output. The Nigerian Government is aware that effective participation in oil production is indispensable in national planning. The Federal government acquired 55% (14) of the stock in all major oil producing companies operating in Nigeria. The Nigerian national oil corporation was empowered to monitor and supervise the operation of these foreign firms. The Federal Ministry of Petroleum and Energy Resources has the responsibility of regulating all oil drilling and the issuing of per- mits. In 1976 the National Oil Corporation and the Federal Ministry of Petroleum and Energy were merged to resolve conflicts of authority and to ensure consolidation of efforts. This new body in charge of Nigeria's oil production is known as Nigerian National Petroleum Development Corporation (NNDC). The creation of this organization is significant for two major reasons: - It emphasized the fact that Nigeria intends to take control of this major economic sector. . It also represents a major effort by a developing country to develop indigenous skilled manpower to meet the demands of such an intensive technology and high capital industry. Nigeria, like most other countries of the OPEC, has the problem of identifying the extent of its oil reserve. An expert predicted 26 that Nigerian reserves may last for (14-16) years (15) at the present production of 3 million barrels per day. It is not easy to predict Nigeria's reserves because 50% of the country is unmapped for mineral analysis. The fact that oil was discovered around Lake Chad basin in an indicator that oil may be found in Northern Nigeria. The areas adjacent to Lake Chad basin has the same geographical fault as parts of Northern Nigeria. As a result of the 1974 oil crisis the price per barrel jumped from $4.29 to $14.69. This increased the contribu- tion of the oil sector to 80% (16) of the total foreign exchange earnings. An additional $500 million dollars was generated in 1974. Higher earnings from petroleum export created a favorable balance of payment as high as N-3.1 billion in 1975. Proceeds from the oil sector provides 75% of the funds proposed for the implementation of the third national development plan. Tables (2-6) and (2-7) emphasize the importance of oil in Nigeria's export trade. In 1972 the oil exports accounted for 87.2% of the export revenue. Agricultural products con- tributed only 10.9%. The dependency on oil has grown to 90% by December 1976. Natural gas plants are not yet available to process Nigeria's gas reserves. Hence 56 (17) million cubit meters of gas are flared per year. This represents a great loss to the national economy. Priority should be given to construction of a natural gas liquifica- tion plan to maximize oil revenue and to conserve irreplacable re- sources. Itappears that port planning should be based on a constant pro- duction of 3 million barrels per day over the next ten years. Governmentzcontrol appears to be strong enough to assume this level of production. 27 .333mm_z .momm3 .m333 goon: .m .oz .3 ._o> .meoumuPuc3 u_so:ouu 33333332 3o 63333333 .mcmumu "mucaom 3.333 3.333 3.333 3.333 .3.333 "33333 3.3 3.3 3.3 3.3 3.3 33o=eeea 3:333: 3.3 3.3 3.3 3.3 3.3 33333323 3e3moeea 3.33 3.33 3.33 3.33 3.33 33o=eee3 332333333233 3.33 3.33 3.33 3.33 3.33 33o=eeea e=o3ee333 ”aaogu 233uoesou >5 mmmucmugmm 3.33 3.33 3.33 . 3.33 3.33 33ee3x3 33333 33 3 333.333.3 333.333.3 333.333 333.333 333.333 ”33333 333.33 333.33 333.33 333.33 333.33 .33: 333333-333 333.3 333.3 333.3 333.3 333.3 333323333 333.33 333.33 333.33 333.33 333.33 3333: 333 - - 3 N3 33 oeo 333 333.3 333.3 333.3 333.33 333.3 333: 3:3 3333 .233333 333.333 333.333 333.333 333.333 333.333 333333-333 333.3 333.33 333.33 333 33 333.33 333333 333 333.3 333.3 333 333 333 3333 333.33 333.33 333.33 333.33 333.33 3333233 5333 333 333.33 333.33 333.3 333.3 333333 333 333.33 333.33 333.33 333.33 333.33 3333333333 333.333 333.333 333.333 333.333 333.333 eooeo 333.333.3 333.333 333.333 333.333 333.33 33333323 233332333 3333 3333 3333 3333 3333 3333. .23 mumhmoozzou 4 .33o33u3333 u3socoum 33333332 3o uF—naqmm 3333333 ”mogzom 3333333>3 yo: .3.: 3333 3333 333; :3 33333 «3353333 A V 3333 .333333 .333333 3333 33 33 33 .3.3 .3.3 .3.: .3.: 3ummmmuoga 3ouou 333V 33 33 33 .3.: .3.: .3.: 33 33:2 33:3 333 3 m m .3.= .3.: .3.= 3 mmmmou 3 m 3 N3 N3 33 33 cm 3303333 .333233 33 cm 33 33 33 33 33 33 swans: O3 33 33 33 .3.3 .3.3 .3.: 33 3333 couuou 3 33 mm 33 33 33 33 33 333 couuou 3 3 3 Q3 3 m w 3 3:333 3:3 3333: 333 3 N3 33 33 m 33 33 ummm 3:333 3333 om 33 33 .3.3 .3.3 .3.3 .3.3 Pmcgmx 5333 N o3 m m m 33 333 333 33c 2333 333 333 333 333 mom _ 333 333 333 mouou 333 333 333 333 333 333 333 333 333333-333 333 33 333 333 333 333 333 333 @333 333333333 33 33 cm 303 333 33 333 33 330 “saunaogu 333 333 ._ .333 333 333 333 333 cum muzccczogu 3333 3333 o333 3333 3333 3333 3333 3333 33333 333.3 3333333 333333333333 .33-33 33333 29 2.1 (c) (iii) Other Sectors: The manufacturing sector has been growing at the rate of 5% (18) per year. The emphasis is on establishing secondary industries for the processing of agricultural products and manufacture of durable goods. Two car assembly plants were established in Lagos and Kaduna under a joint participation program between Volks- wagen of Germany and Pegeout of France. However these plants are of intermediate scale and jointly produce only 5% of the total automobile demand. Four major cement factories are located in Sokoto, Nkalagu, Calabar and Ewekoro. These factories jointly produce 40% of the total cement used in the country. At present there is no major National Iron and Steel Industry; the Nigerian steel authority is still at the re- cruiting and planning stage. Hence like most developing countries Nigeria relies heavily on foreign countries for the supply of cars, equipment and construction materials. The manufacturing sector is lagging behind the GDP which is growing at 10%. This situation is due to the lack of sufficient capital in industries. Government spending alone will not be sufficient to achieve an adequate growth level in the industrial sector. The second major problem is the shortage of intermediate technicians and engineers. Government services is the highest growth area. This sector is growing at an annual rate of 30%. (19) The government policy is to provide better health and educational services to the people. In pursuance ‘TF these objectives universal primary education (UPE) was introduced in 1976. Physical infrastructures to accommodate this program doubled government spending in education to N-779,362,610 in Table (2-8). 30 Ministries/Departments State House/Dodan Barracks Cabinet Office Police Police Service Commission Agriculture, rural development Audit Aviation Co-operatives and supply Communications Defence Economic development and reconstruction Education Establishments and service matters External affairs Finance Health Industries Information Internal affairs Judicial Justice Labour Mines and power National science and technology Development agency Nigerian National Petroleum Corporation Public Complaints Commission Public Service Commission- Trade Transport Nater resources Works Federal Electoral Commission Non-statutory appropriations Consolidated Revenue Fund Contingencies TOTAL ALLOCATIONS UNDER RECURRENT BUDGET Allocation N-1,440,510 N-41,631.210 N-127,625,850 N- 141,420 N- 19,711,170 N-1,360,000 N- 20,608,620 N- 2,225,400 N- 380,500 44- 597,857,007 N- 27,714,430 N-779,362,610 N- 19 ,332 ,280 N-32,589.990 463,060 ,920,000 *1: 43H 54,002,217 N-28,777,296 N-2,800,000,000 His Excellency General Obasanjo: Budget Speech 1978. Federal Ministry of Information, Lagos, Nigeria. 31 1978. Hence education was allocated 28% of the recurrent national budget, making it the largest government service activity. Figure (2-1) illustrates recurrent government spending in major service areas. Defense and police account for 26% of the recurrent budget thereby ranking second to education. Health and Works services have 2.9% and 3.3% of the total budget funds. Agriculture which is the second dominant economic sector was allocated only .7% of the recur- rent budget. Industry is the lowest on the ladder with .12% of the recurrent budget. 2.1 (c). Population In 1952, 1962 and 1963 population census were taken in Nigeria. Table (2-9) indicates the national population estimates between 1963 and 1968. (20) The 1962 population figures were controversial because of allegations of irregularities. Hence in 1963 Nigeria had to go through the costly exercises of a second count. When 1952 figures are matched with 1963 figures, a growth rate of 6.2% per year is obtained. This growth rate is not credible because it is far higher than the United Nations estimates of population growth for Africa (2.5-3%). This error is due to the fact that the 1952 figures were undercounted while the 1963 figures were high due to double counting in some parts of the country. United Nations statistics put the growth rate of Nigeria's popula- tion at 2.5% per annum. The Federal Ministry of Transport developed a reliable forecast of Nigerian population based on UN and National Institute of Social and Economic Research (NISER) growth indices. Table (2-9) indicates that Nigerian population in 1985 will reach 96.5 million. 32 INDUSTRIES ’////.12% TRANSPORT ADMINISTRATION .3% AVIATION .74% AGRICULUTRE EDUCATION 2.8% .72 OTHER SERVICES 4.6% DEFENSE Figure (2-1). % ALLOCATION UNDER RECURRENT BUDGET 1978. Prepared with data from National Budget 1978 (see Table 2-8). 33 Hmm.om mfi~.mm efim.mw mmmw m.~ m.H m.¢ .mmummmw Hem.mm www.mo mm~.- mmmw .mpcmmwz .mommJ .mccouwm umsmrpazmca "Po: u_:: mcwccmpa ugoamcmgh "muczom m.~ m.~ m.~ m.~ Am xv .Aofi-~v apnmh "ho: avg: mcpccmpa agoqcmch "mugzom mom.mm Nuo.mm 4wm cum mmm mom mum New mom.m~ pmsuamu ummm 00“ mm mm mm . om weo.wm :Lmummznu_z umm.~ ~¢o.~ mam Nam cow mmm.m moan; «mm mom omH mmH mma mom.mm :smummz mm Ne Nv mm mm omm.¢m mgmzx mm Hm me ow em cm~.mm~ cgmummz gugoz mm mm ~m me oe www.oo~ ammumpm macmm me we mm em mm cmn.mmm cgmummm cugoz mmmH gmHWw—W." 3. mLW%WW Lmn— MfiWMLmn— mmqm Eumbm mmHhHmzmo onh<43aoa .Afiuumv mpnmh 36 network. Modern economic theory holds that new transportation systems induce flow of goods and services which otherwise would not occur. (21) In developing countries such as Nigeria the need for efficient trans- portation systems is even greater due to the rapid growth of the economy (10% per annum). In the Third National Development Plan the Federal Government of Nigeria acknowledged the importance of efficient transportation systems in these terms: "Nigeria's transportation objectives have been stated since early 1960, in general terms as aimed at co-ordinated development, economic efficiency and, by implication, the support of national interests like the opening up and binding together of this vast nation. These objectives are as relevant and valid today as they were when first explicitly set down in 1965." In pursuance of these objectives the Second National Development Plan emphasized transportation capital investments. Between 1970- 1974'N-485.2 (22) million or 23.7% of total public investment was injected into the transportation sector annually. This expenditure summed up to N-1,025.4 million by the end of the second development period. By 1974 capital expenditures in transportation topped the list fbllowed by other service areas like education (13.5%), agri- culture (10.5%) and health (5.2%). Public Investments bngodes Table (2-12A) and Figure (2-2) illustrate the proportion of public investment allocated to each mode. Roads took the lion's share with 68.5% of the total capital investment. Railroads had only 9.0% while ports and airways got 7.4% and 10.6% respectively. An analysis of each of the modes is essential to assess the strengths and weaknesses of the above investment policy. 37 HIGHWAYS 68.5% AVIATION AND AIRWAYS i 10.6% . B (til/AP qu$X§gl .{9 Sb . Q) V~ ‘3 o". V§§ ‘<§$$\ ‘0 $3 :3 3 N < a .9 ea. .105 2": Q: TOTAL INVESTMENT IN TRANSPORTATION BY FEDERAL AND STATE GOVERNMENTS OF NIGERIA, 1974. Figure (2-2). prepared with data from Table (2-12) 38 .oumfi .mommg .copumsgowcm mo acumwc_z Pocono; .cnmfiuomma :mpa acmsaopw>wo Pacowumz acoumm ox pt 06 .4 tom O H m—nxo Hooo q-FOF3 fijFfiFfl acmsumm>=H we cowagoqoga m.om~ m.¢mm N.mm¢ «Hm m.m~ o.n~ o.mm n.m¢ m.e¢~ m.an c.~mm 5:522. HcmEHmm>zm $0 u::OE< mathm hmommzzm enuoumm mhzmzzmm>ow 44< >m "m_gmmvz mo uppnznma .msmumu "moczom muhgmm mewupgmz mcvaawzm mmumum ~_<-- pacmummTT "pmuop-- ”mangmumz ccmpcn mxmch< cowum_>< pp>Pu wagon amzp_mm mauapm ._<-- Pmcmcmmuu "Pmuohu- "mumoa S<3.=<¢ II I I I oo<0¢ ou>... ...l.l.l 33323.. 3:... II... nus-.323: 45.9.2255. 0:99 . «2.523: ozuwwd g o. 00 08 OCUbUIO: i Ugo” I”. fight . 2.x .. . c u . .9. " esxfiyfi. A. o l.:\. <28: .3 .2: 5:52sz .8.3 2.5: 41 Police. Regulation however is limited to safety requirements. Fare setting and route allocation are completely ignored. The result of this incomplete regulation is arbitrary setting of prices by individual operators. Movement of goods and services are hindered by unreasonable fares set by owner operators. Agricultural products are affected most because the cost of transportation absorbs the farmer's profit. It is evident that efficient flow of goods and services cannot be achieved by providing only good highway systems. Economic regulation of carriers is as important as physical infrastructures. Nigeria Federal officials should consider fare and route regulation as en- . forced by the ICC in the United States. 2.2 (c) Nigerian Railway,Corporation As illustrated in Figure (2-4), Nigeria has 3,505 Km (26) of single rail track linking the two principal ports of Port Harcourt and Lagos. These tracks still retain the traditional 3'-6" guage (27) which were predominant, in many parts of the world, during the 1920's when these tracks were constructed. In fact the railways are outdated in terms of equipment, tracks and capacity. The track curvatures are numerous and in many locations bridges are old. The Railroad still relies on traditional signal systems and there are no automatic switch- ing systems in the country. Scheduling is irregular because of rampant breakdowns and unavailability of both movement and power units. The speed of delivery is very low because of track conditions and the age of the equipment. Trains average 40 Kms per hour, with fre- quent stops and unnecessary delay. As a result of these inadequacies, the Nigeria shippers prefer the trucks to rail service. As indicated 42 Figure (2-4). RAIL NETWORK IN NIGERIA WESTERN DAHOMEY EASTERN BIGHT OF 200 BENIN ‘VB RIVERS "calabarn 03 #1 ~( (Wé: miles «,0 «5 arcour JV“? _ BIGHT 0F BIAFRA kg ‘ Major Cities . Major Sea Ports NNH‘ Rail Lines 43 by table (2-128) the Railroad traffic declined in both tonnage and monthly average length of haul between 1963 and 1972. Consequently the annual revenue declined from-N-32.6 million in the 1963/64 year to 24.5 in the 1971/72 year (i.e. a decline of 25% in 9 years). Since 1950 the Federal Government has stablized the railroad by providing both capital and operating subsidies. In 1972 NRC spent more than its revenues by 50%. With increasing costs incurred in other govern- mental services like education, works and health it is questionable whether the Federal government can continue to provide sufficient subsidy to the NRC. . In conclusion, the Nigerian railroad faces serious problems. The NRC runs a deficit each year in the face of sharp freight compe- tition from the trucks. In 1958 850,000 tonnes of farm products moved by rail; in 1970 this declined to 350,000 tonnes (28) (indica- ting a loss of 59%). In 1961 passenger traffic stood at 11,000 per day; in 1974 it dipped to 4670 per day (indicating a drop of 58%). Operational deficits have been increasing. In 1973 the NRC losses totalled-N-21.8 million as against N-33.1 million in 1974. (29) It is apparent that there is a need for the Federal government to in- ject funds into the NRC to phase out outdated 3'-6" track guages and purchase new rolling stock and power units. A new management is also needed to introduce modern techniques like programming, system schedul- ing and traffic co-ordination. The NRC needs to adopt a marketing orientation to ward off increasing competition by truckers. Improved shipper information and increased speed of delivery are essential in this direction. 44 NmN.H~ .mNH.m .me me ow “mm Hm Rafi.“ m.¢~ ---mfl omm.- www.mm ~.NH mm mm mam com ¢m~.m c.~m ecummmfi m_cmmwz .mommA .:o_umgoagou xmzpwmm :mwcmmpz ”wuczom Pwmmpo Emmum "mes—wad mcpmcw 2mg .smx mcpmcu mama n::om-:s=b comm: mamgm>< owemmgp cw some: Lma xmv can .max mmmgm>< xuopm cw comm: Lon Asa gun .msx mmmcm>< “Exv Pam: mo spaced mmmgm>< »_gp:oz Aooo.v wmoccoh mcwxamucoz mmmgm>< mpgucoz Aooo.v ommccoh mcpxma mmmgm>< xpnucoz Acowppps.zg Lama go; macm>mz muz<=SH<¢ ‘ I RAPIDS §Kadun( g :. JEBBA ~ (01°53 , BENUE R. 8 GOYA‘ { BRIDGE . 0) Y0 LA 8 MAK ROI BRIDG ‘ . I 63H" ‘ Bi. ' E IlLATEA - § .- . .2.» Ibadan - " N ' WESTERN - i 1160 D“ E8 Lagos openin°c 51:?891‘“ rASTz RIGHT OF “5512' BONNY9SRTIVI‘AO BENIN > v- S Gealab‘"; FARCADOS q ‘ '" miles arCOUrt RIVER IGHT OF BIAFRA 47 Niger widens to 2 miles breadth and has a draught of 30'-50' (30) in the rainy season (June-July). The Niger continues its journey to the sea by breaking up into several small rivers. The main stream continues until it reaches the delta where it divides into fourteen main outlets to the sea as shown in fig. (2-5). The more important outlets are the Bonny River and the Farcardos River. The Bonny River opens up a waterway which services the special- ized oil port of Bonny. The Farcardos River is the gateway to the ports of Warri and Burutu. As illustrated in figure (2-5) River Niger is navigable for river steamers up to Jebba. After Jebba there are problems of falls and rapids. Tugs of 1500 HP pull 3-4 barges of 500 ton as far as Goya and take only one 500 ton barge through the Awuru rapids. (31) Hence the Awuru rapids is the main impediment to navigation on the upper Niger;' with depths as low as 10-15 feet (32) and tugs run the risk of running aground. After Onitsha the lower Niger is navigable for the entire year. The draught is between 20-25 feet (33) until it divides into 14 delta rivers. Tugs of 1500 HP can effectively pull 3—5 barges of 500 tons. The main tributaries, the Bonny and Farcardos Rivers, are open all year. These rivers combined with coastal creeks are capable of taking 20,000 ton ships and medium size ocean tankers. However the extent of navigation is limited to short channels from the sea to the Warri River. Traffic on the Niger River has been declining due to the lack of efficient equipment and funds required to revitalize inland water ser- vices. A River Transport Corporation was created in 1973 to investigate 48 the feasibility of year-round navigation on the Niger river. This corporation was also empowered to operate vessels on the Niger. The corporation failed to deploy a single vessel on the Niger in a period of 5 years. It was a creation of three states and as such administra- tion was difficult. The other lower Niger States declined to invest in such a venture. This situation is an indication of the fact that development of the Niger river cannot be handled by a group of states. It will require a Federal initiative to open this major waterway. In recent times dams have been constructed at several points on the Niger. In Senegal a hydro-electric dam is under construction while in Nigeria the giant Kainji hydro dam was completed in 1968. These dams do not leave sufficient compensation water for navigation on the Niger. Benue River Traffic on the Benue has also declined. A number of factors con- tribute to this trend. First there are major impediments to navigation as indicated by table (2-13). Navigation up to Yola is possible for 3% months of the year for vessels with draft of 2% m. In the first 2 months draft of 4 m are available. Even during the navigation season temporary drops in water level delay navigation between Markudi and Yola. Secondly the channel is tortuous and rugged. As a result of this the length of tow is limited to 107 m maximum for single barges, 12 m wide or 76 m maximum length for double barges 23 m wide. Table (2-14) indicates the total tonnage of traffic which the Benue River handled between 1950-1960. In 1950, at its peak, 50,000 tons were handled by all Benue river ports. In 1973, twenty-three years .HHmH .mommH .Hgoamcmgp mo agHchHz ommHuoan upcmmwz we mason mzu mo Hcmsaon>mo "muczom o.~ mm HHco umuaum weaves N cog» gmummgm mugmzaso \ weaves m.H vac ~.H cmmzumn mmpcm> pnmzmgu mymm ".mww 49 N H wHammH>mc no: u c "mmuoz com.H macaw .mwp .mww cum o.~ m.H m.H c : m.H m.H m.H m.H ~.H N.H c c c : m~¢.H mHo> .mwp .mnw .mww .mww : : m.H m.H o.~ o.~ m.H m.H N.H : c : Hmm Hugaxmz .mwp mww .mww N.H m.H o.~ o.~ o.~ o.~ o.m o.~ o.~ o.~ m.H N.H mucmancoo mH H mH H mH H mH H mH H mH H .>oz .Huo .Hawm Hmzma< HHza wcaq Ex :oHumooH .mm>Hm mazmm uzh zo zomm mhmom zmo zomzH19 Lacuna: Pacommmm m.mm m.om «.om “.mm H.5m «.mm H.5m ~.Hm ~.mm H.~m m.mm m.mm Papa» o.om m.m¢ N.cm m.me m.o¢ m.m¢ m.o¢ ~.H¢ m.ov «.mm m.mm H.1e mggoasH m.m¢ o.H¢ ~.mm ¢.m¢ m.m¢ H.N¢ ~.o¢ m.om N.m¢ ¢.me H.mm ~.mm mggogxu ommgm>< m.m~ m.mn ~.Hm m.mc m.mo N.me o.mm N.mm ~.Hm o.Hm ¢.~m w.mm mugoasH m.mn o.mu o.m~ m.¢u o.o~ o.mc m.o~ o.¢m ~.mn m.c~ ¢.nm o.Ho mugogxm omoH o.oe o.mm m.me m.m¢ o.m¢ m.ev N.He ~.H¢ N.~¢ m.mv o.¢¢ m.om maHoQEH o.m¢ m.~m N.mm m.om «.mv m.¢e m.mm o.o¢ o.~¢ m.~m N.Hm N.~m mugoaxu mmmH m.H¢ m.o¢ N.oe m.¢m o.mm «.mm m.om m.~m «.mm o.mm o.n~ ¢.Hm mugoasH «.mm e.om m.w~ m.w~ m.m~ m.m~ m.Om o.mm o.mm m.em o.H~ c.om mugoaxu momH o.mm ¢.~¢ o.mm o.mm o.mm o.mm o.mm m.vm ¢.Hm e.mm o.~m e.mm masonsw o.m~ «.mm o.nH m.m~ m.e~ o.o~ c.m~ o.mm o.m~ m.- o.om o.m~ mugoqu ommH .uoa .>oz .uuo .uamm .m=< xpza mczq Am: ~_La< .Lmz .amm .cma mgmm> AcowP—wz .zv mo H .09 .’. Xo > X2. Hence there is no significant import variations during the dry and rainy seasons. (ii) Applying X2 test on seasonal export data: Dry Season Rainy Season 0i-E 2 Export Average Export Average H4 nfillion) (N- million) 26.0 27.8 .12 30.7 29.4 .05 47.2 48.9 .06 71.6 73.2 .03 01-5 2 E = .262 . X3 05 confidence level with degree of freedom 3, x = .352 > x2 > .262 o ’. there is no significant variation between the volume of dry and rainy season exports. 69 Hence there is no significant seasonal variations during the dry and rainy season. The result of the X2 test is contrary to general expectation thatLthe rainy season affects export and import activities. This analysis con- firms that any disruption in these activities (e.g. loading and unload- ing of ships) is merely temporary and of no signficance in affecting aggregate seasonal export and import levels. 7O m<2 O_In_322... .233. 3:32.. 2.2.: 3.33 3:. . ooo.ov.ooo.o~ . ....a...ooo.ao - 3......ooo62 I 3.636363 E :26 92 28.8. zo_h<..5n_on_ 24mm: 2.2:... no: go 323 © on 525 mmmmm 3.3 3...: :72: ...... .353 ll :26 2.4 a: 3;. 43:“. .Ex wm._<3:@=* 2.3.8.. .33) 8 33...... .33.: 3.3.... 71 3 OCH-332 It: tobtt n»<3.:<3 'II! b4’0‘08 Chaos: I'll-O O§IUhIU H....nug . Ill C040. .8.~n.lu . acuzt . Ill '.' .284; u~m:w¢ .om_ ocoa co Pamsasma< "meowumz umppcz "mugaom Xao m_ .uumwwm cow—apapss a zmaoczu muwmwcmn cw mmmmsucw A_P_v momsumzucw vmumpmciucoa cu msoucw cw mmmmcucw AP_V conmp toumpmgiucoa on usage? :_ mmcmgucw Apv mgopumu u:n:~ mo mgmwpaazm cu mupmmcmm pumcpccm acmsumm>=w axon an mpnwmmoa mums zgumaucw com: -ugoa mo uzauzo :P mmmmcucp unusumm>c_ econ an xpawmmoq mums mapgm gmmgmp mcwwmcmqo mo opmom we asocoom soc» mcwm_cm umou mcvamcmao m.aw=m =_ mmcw>mm ucoa cw umou .mawcm cw mm=p>mm >gou:m>:_ cw a: new» Pmuwnau we mmcmaxm pmwcopcw one :_ mm:_>mm umoo mocmcam=_ cw mm:_>om umoo mcvpucmsiomcmu :_ mm:_>mm amoo “commence ncmpc_ :w mcmm: ugoo op wewkmcmm > "- mmcw>mm Apv pcoEpmm>cp pumnoca on» an opnpmmoq mums vamp Go Faucoc Pm:o_u_num APPPV m==m>mg mew—ccms -omcmu am: cm mmmmcucm Aw_v ma_;m so most Easy mzcm>oc pacopu_uum APV osoa o» u_co=om ooos_a H2m2hmm>z~ Exec no mhmumzum mammmmca .AH-mV spank 87 costs) in the study of the port of Casablanca. This is one of the reasons the literature on the simulation of general cargo ports is limited. Most port planners simulated bulk ports which are less com- plex than general cargo ports. Economic models can be applied to the evaluation of alternative port investments. Their main weakness lies in the definition of in- tangible benefits. This area tends to be subjective. In addition in most developing countries ports are state owned and emphasis is on thorough-put rather than investment pay-back considerations. Economic models are in some cases simplistic and ignore the upward multiplier effect on the economy which an efficient port creates in the hinterland. Analytic, simulation and economic models each have a role to play in port studies. The results of analytic models can provide input into complex port simulation models. Port planners need to employ analytic models to calibrate the simulation models. The results of the simula- tion models, in turn, become inputs into the economic models employed to consider specific improvements. Hence there is still tremendous work to be carried out in the areas of port analyses and development of investment criteria for general cargo ports. This need is even greater in developing countries where a rapid economic growth rate and higher level of imports are creating congestion in the existing port systems. 3.2 (b) Methodology. Ship traffic through the port system can be analyzed by the application of simulation model of multiple service channel type. Figure (3-3) illustrates the general configuration of such a service system. There are three major assumptions: (18) 88 Figure (3-3). CONFIGURATION OF SHIP QUEUING SYSTEM ('SINGLE WAITING LINE' DEFINED BY ARRIVAL TIME AND PORT PRIORITY POLICY) __1. i E l E ZONE MARSHALL I NG L :L’[ L ! l .1. ._ j 1 \___/ //\i:/ / l I J K! j... SHIP AT \ HEAD OF OUEUE n / I—..__‘=.-Lh-__—- I I . N BERTHS TRANSIT . WAREHOUSE TRANSIT WAREHOUSE TRANSIT WAREHOUSE I L1 #1 ii / ..... .i f7“: 89 (i) A 'first-come first-serve' queue discipline is assumed. (ii) All berths have the same service rate (and service capacity). (iii) Ships do not line up behind one another as they arrive but are held back at the anchorage until a vacancy occurs. The first assumption is consistent with policies used in major world ports, e.g. Lisbon, Casablanca, Bangkok, and London. (19) However ports do have priority policies for express vessels, military vessels and tankers (in multi-purpose ports). In such a situation a 'first- come first-serve' discipline can be distorted if the volume and fre- quency of arrival of the priority class is significant. When this occurs preemptive-resume priority statistics is applied to the simula- tion model. (20) In this technique a class A ship will be served upon arrival unless there is another class A ship already in the berth. Thus in a multiple berth port a class A ship will preempt a class B ship al- ready in one of the berths. The class B ship automatically returns to the head of the queue waiting for the next gap. In practice this situa- tion is not easily carried out because of the difficulties and time cost of unberthing a half off-loaded vessel. The priority ship merely moves to the next vacant berth in some ports. (21) The second assumption should be discussed in relation to specific ports. In some general cargo ports, with deep natural harbors, berths accommodate all classes of general cargo ships. Cranes and transfer equipment are often drawn from a common pool serving all berths. Thus the second assumption should be considered with conditions in specific ports. The third assumption is in line with existing Harbor master's policy of marshalling vessels. As a ship arrives at the anchorage it 90 joins the queue and waits until it is at the head of the queue. Figure (3-3) illustrates the multiple channel structure of a general cargo port. It is important to note that even though ships cluster around the anchorage a queue still exists. This queue is based on time of arrival coupled with the port priority policy. The capacity of N ships in the system at any given time is a function of both the physical limi- tations such as the number of berths available and the level of waiting that is intolerable to arriving ships. With general cargo ships balking rarely occurs because of the high cost of ship operations. This implies that the arriving ship has no choice but to join the queue. This situation is more common where alternate ports are not near the desti- nation ports,~e.g. Lagos. As a result of the above discussion the population of ships are considered infinite. The (M/M/C):(GD/N/m) (22) is considered an ef- fective model in dealing with infinite queueing problems. An important assumption is that a single waiting line exists; jockeying and reneging do not apply to ships because of strict harbor control policies. A detailed discussion of the model is given below. Let Pn(t) denote the probability of n units in the system at time t Let A(n) and u" represent arrival and service rates when n units are in the system: then ant = probability of one arrival during At 1 - AnAt = probability of no arrival during At unAt = probability of one service during At 1 - unAt = probability of no service during At Ihithe case where At is sufficiently small that both an arrival and a service can not occur during At: 91 Pn(t + At) Pn(t)(1 - nxt)(1 - unAt) + Pn_1(t)xn_1 t( 1 - un_1At) Pn+1(t)( 1 - xn+1At)un+1At . . . . Equation (A) + From equation A above the probability of n units in the system at any time period is the sum of the probabilities of the following three events: - n units in the system at t, no arrival or service during At - n-l units in the system at t, one arrival and no service during At - n+1 units in the system at t, no arrival and one service during At As At + O and dividing equation A by At we obtain dP (t) - 2 T 'Pn(t)(xn + 11n) + Pn-1(t)“n-1 + Pn+1(t)“n+1 ' ' Equation (8) At steady state t + m or.dPn(t)/dt = O In steady state Pn(t) is not a function of t, i.e. t can be excluded and equation 8 above becomes: 0 = -Pn(xn + “n) + Pn_1xn_1 + P for 0 < R < N . . . . n+1“n+1 Equation (8.1) O = 'Poxo + Plu1 , O = PNuN + PN-IAN-l . . . . Equation (8.2) Equation 8.1 and 8.2 are the general balance equation of the M/M/C queue. We can solve the general balance equations as shown for n = 1, from Equation 8.1 P2 = A1 +“1P1 - AoPo , solving for HZ U2 _ A A P P2 - 1 o o Uzul 92 n}. II k-lPoforliniN Generalizing P = " k=1 “k r' N n but Po - [-1 + n= :11 k-1 , .. for n - 1, 2, . . . . N . Equation B.3 Table (3-2) shows all the formula deduced from the M/M/C steady state general balance equation. Note that the average time spent in the queue and the average waiting time can be determined provided that the utilization factor 0 < 1 (where p = A/ku , A = arrival rate of ships, k = number of service channels, u = mean service rate). When a > 1 the M/M/C steady state model breaks down. This is a major dis- advantage of the above model. In addition the M/M/C steady state equations do not consider the dynamic nature of port operations. Hence in this dissertation a simulation model will be developed. 3.2 (c) Optimization. The objective of the optimization program is to specify the 'optimal' set of port resource requirements which will minimize total ship and cargo delay given a defined demand situa- tion. These resource requirements include all subsystems (e.g. berths, cranes, warehouse, storage areas and transfer equipment). Table (3-3) illustrates the UNCTAD definition of the operational subsystems for the port of Casablanca. In a port system analysis the major objective functions may be Table (3—2). Source: M/M/C STEADY STATE EQUATIONS flfultiplc-station queuing relationships with Poisson arrivals, exponential service-times, and leading vehicle in queue moving to first vacant station‘ for steady-state conditions Queuing model Description of model I A " p(n) = probability of having 1 p(n) n! (:1) 17(0), exactly it vehicles in forn=0,l,...,k—l systemior0_<_n_ k) P(d _<_ t) =- probability of 8 PM S t) - l — e l {1 + 7— having spent time X 1 _. e-uuix-muki-(mn t or less in system 1 — (Milk) - (l/k) if . A .g ”(0) P(n 2 k) = probability 0‘ 9 1’0! 2 k) = 2 PM) = (:4) -———A— having to wait in a-k kg (1 _ .1.) queue p . ‘ k - number of service stations or service channels, each having a servicing rate p M - mean arrival rate per station (as used in Example 11.] and Table 11.8). x -k)u. Wohl and Martin: and Planners. McGraw-Hill, 93 Traffic Systems Analysis for Engineers New York. 94 Amxgozao: Pam; woo woo; mcaozpocav Loam .N Aomcou pogocomv coma oz .a mama; N swan mgzaam . . .m Amxcozaoc pang ooo coo; moans—ooav Loam .N Aomcoo Fococomv coao.oz .a woman a Loan ocoaad . . .m Aomcoo Amxcozao: accocmmv Loom; Pam; oco woos moaoapuoav Emamzm acommca .a mcoaa pooaoo .mcoao . . .u 5 ma gamma .e 5 aa saooo .m E m gamma .N m xooo acamooco oz .a moamoogo ogoaza .mcamooao . . .o E ma zaooo .v E aa gamma .m a a :aaoo .N N xooo mcamomco oz .a anamooco ocaaza .mcamooco . . .m 5 Ma canoe .e 5 Ha gamma .m s a sand: .N a goon anamooco oz .a mcamooeo ocoaow .mcamoogo . . .q conga; anmzm acomoco .a moamooca acomocq .mcamoogo . . .m zaaon cams oga wo :oamooaxo oco Ammoam aocawv zaaon 3oz .m Ammoam amcawv zaamm 3oz .N Amoaaaon ozav soamzm acommco .a mcoaozxoosm mgoaozxoocm . . .N meoaozxooao 3o: gaaz anmzm .N anmzm acommca .a Emamzm pocmam Emamzm anomam . . .a - axon Emamzm amass: coaaaoaaoo osooooau no mo»» ooaagacomoo acoo Emamzm m mo monthHmmo cmm .e mooacon xoon uoaooo zazo :o mmna xam .m mooaamn xoon moagzo zaao :o mmoa o>an .N Aacomocn ao anmzmv mooaaon goon mcagzo zaoo Eo mmoa soon .a omozop zoco oco was» . . . .aa mooacmn goon moagoo zaoo co maopan xam .e mooaoon xoon moasoo zaoo co maoaan m>an .m Maommocn ao anmzmv Aamogu mooaoon xoon moaozo azo co maoaan coon .N o goo maoaan ozav mooaaon zoon unaczo zaoo co maoaan omega .a omoaoaan maaonu oco maoaan . . . .OH anon anmzm anaacaooo ogooooam mo onza coaanacumon LooEzz anon anmzm ooscaaeoo .zm-mn manna 96 .nnma .oanmEm .xaoanoz .oam .oEooznm a ozom .oanomaz Enoooz “zoo oaoEn< ”oonzom NE ooo.m .m NE oom.e .N ouonm Manmonn aov NE ooo.m .a omonoam ooaonomanaoz . . . .mN ooo m oomoao ano non mm .e ooonm oomoao anu non om .m ouonm ommoao ano non mN .N N nman ooonm ommoao ano non o .a omonoam onzaoa .omonoam . . . .NN moonm oomoau anu non ma .e moonm oomoao ano non om .m muonm oomoao ano non mN .N a noan moonm oomoao ano non o .a maonoam onaaoa .omonoam . . . .aN noonoE anmonn zoooa anmzm .a omonoam Ea mmonoam . . . .oN aaaEm non mmEom OON .c aaaEm non mmEom oma .m aaazm non mmEom sea .N anE Aanmonn aov aaaEm non moEom mm .a inazco mEaaoEo: mmEoo . . . .ma aom: Ea ooNv maaE: noN .e Aoma Ea Oman maaE: ooN .m aom: Ea ceav maaE= ona .N anE mnoaaona Aanmzm anmmnnv Aom: Ea mmv maaE: QNa .a -naocm mEaaoEoz oEo mxoona . . . .ma ammo Ea omav moEonu OON .e zoo: Ea naav moEono oma .m Aanmzm anmonnv ammo Ea Nov moEono OOH .N anE moEono oz .a -naoco mEaaoEo: moEonu . . . .Na anaaEaaoo onooanm anon anmzm anananumon nooEEE ao onza anon anmzw. umzcaacou .Amimv oanoa 97 cost, delay or thoroughput. Cost and delay are minimized for a spec- ified thoroughput. One example of a measure of port performance might be: Minimize Z = aX + bX3 + dX + ex5 + + + + 1 4 (Xe/T6 + Cv6) + (x7/T7 + CV7) + (XS/T8 + CV8) (x9/T9 + CV9) + (XIO/TIO + CV10) + (x11/711 + CV11) (x12/T12 + CV12) * (x13/T13 I CV13) * (x14/T14 * CV14) (X15/T15 + CV15) + (X16/T16 l CV16) + (X17/717 + CV17) + (XIB/TIB + CV18) + cx19 + CX20 where where xi Ti Cv. Z 1 th investment th capital cost of the i economic service life of the i investment annual variable cost associated with ith investment total annual cost of port operation and ownership average cost of one ship/unit waiting time estimate of the average ship waiting time in the queue average cost of one ship/unit berthing time estimate of average ship berthing time average cost of one cargo unit waiting time in transit warehouse estimate of average cargo waiting time in the transit warehouse average cost of one cargo unit waiting time within the inner harbor facilities estimate of average cargo waiting time within the inner harbor facilities land allocation investment for anchorage facilities life period of the anchorage investment 14 ><><><-|><-t><—i><-l><-i>< NHHHHHHi—or—IHH otommNNONO‘U‘IU‘I-h N ...; 98 tug investment life period of tug investment berth investments life period of berth investment Gantry crane investment life period of the crane investment land allocation investment for open storage paved area life period of the open storage investment land transportation investment (unit trains) life period of train investments storage warehouse investment life period of the investment transit warehouse investment life period of warehouse investment dredging investment life period of dredging investment signal system investment life period of signal system investment handling equipment investment (fork lifts) life period of lifters yard transfer equipment life period of transfer equipment floating crane investment life period of floating crane investment number of labor gangs number of supervisory staff average unloading service time for ship 99- All of the above variables may not be relevant in a specific port situation. Variables which are not relevant may be ignored. A discussion of linear programming theory is a prerequisite in evaluating the applicability of the above technique to port performance evaluation. Geometrically linear programming relationships are defined by straight lines in two dimensions, planes in three dimensions and by hyperplane in higher dimensions. These relationships are in the form: (23) A1X1+A2X2+A3X3+ . . .. AjXJ._<_b where Aj's and b are known co-efficients and Xj's are unknown variables. In general linear programming models consist of simultaneous linear equations which specific conditions of the problem and linear functions which define the objective of the problem. The general setting of a linear programming model is as follows: (24) Maximize .0. z=_Z_c H J'XJ' Subject to: n ..X. b' . O XJ 3_ where i = 1, 2, . . . . , M and j = 1. 2, . . . . n This form of the equation was used to determine optimum investment strategies for Lagos Port. 100 To minimize Z, multiply Cj's by -1. Note that Cj's, Aij's and bi's are inputs. The computer adds slack and artificial variables when a linear programming package is utilized. 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES United Nations: Imgrovement of Port Operations and Connected Facilities. TD/B/C4/42 UNCTAD, Geneva, 1969. Gooneratne, S. G. and Buckley, 0. J.: Operations Research Models for Bulk Handling,Systems at Sea Transport Terminals with Particuier Reference toPort_5embla. Report No. 2 School Traffic Engineering, NSW, 1970. Manetsch and Park: Systems Analysis and Simulation with Applica- tion to Economic and Social Systems. Ports I & II, MSU. East Lansing, MI 48824. Robinson, Ross and Keith Tognetti: Modelling and Port Poliey Decisions: The Interface of Simulation and’Practice. Marine Services Board. New South Wales, Australia. Da Silva, F. M.: Boletin do Porto de Lisbon, No. 150, 1963. Nicholan, Stairos N.: Berth Planning_by Evaluationegf Congestion and Cost. Journal of the Waterways and Harbor Division. Proceedings ASCE, 1967. Robinson and Tognetti, op cit. United Nations, 1969, op cit. Gooneratne and Buckley, op cit. Robinson and Tognetti, op cit. Gooneratne and Buckley, op cit. Mettam, J. D.: Forecasting Delay to Ships in Port. The Dock and Harbor Authority, Vol. XLVII, No. 588, 1967. Orner, Ron: Port Simulation Program. Mettam, op cit. United Nations, 1969, op cit. Gooneratne and Buckley, op cit. United Nations, 1969, op cit. 101 18. 19. 20. 21. 22. 23. 24. 25. 102 Wohl and Martin: Traffic Systems Analysis for Engineers and Planners. McGraw-Hill, New York. United Nations, 1969, op cit. White, Schmidt and Bennett: Anelysis of Qgeueing_Systems, Academic Press, New York. Nigerian Ports Authority: Yearbooks (1970-1977) NPA. 28 Marina, Lagos. White, op cit. Gass, Saul: Linear Programming Methods and Application. McGraw- Hill Book Company, New York. Spivey, Allen and Robert Thrall: Linear Optimization. Holt, and Winston, Inc., New York. Gass, op cit. CHAPTER IV FIELD STUDIES PROCEDURE 4.1 Introduction In the study of general cargo ports, data collection is usually a costly exercise because of the number of logistical subsystems in- volved and the disaggregate nature of cargo traffic through the port. The data required for port studies can come from either primary data or historical data. Primary data is ideal but direct measurements and recording require a large number of trained staff, extensive equipment (i.e. boats for harbor and queue reconiassance, automatic recorders, etc.) and substantial budget. In ports with significant seasonal demand variations the duration of primary data collection could be as long as twelve months. Hence labor and other costs in- volved in primary data collection in a general cargo port can be very high. In this study historical data will be utilized in combination with extensive surveillance of the entire port complex. The reasons for employing historical data are summarized below: - The budget for this research is limited and cannot meet the high cost of primary data collection. - In the shipping industry in West Africa, ship and cargo traffic tend to follow a historical pattern due to the fact that the Conference lines have an established route schedule. - The port of Lagos has an organized department of port statistics with statisticalassistants attached to dif- ferent port subsystems. Hence historical data published in the daily records are considered reliable. 103 104 However there are major disadvantages in utilizing historical or secondary data: the existing recording format may not be as comprehensive as the planner may desire, or the logistical flow sequence of ship and cargo may not be arranged in correct order. The author relied greatly on discussions, tours and boat surveillance for an evaluation of the validity of some of the historical data collected in Lagos port. Physical reconaissance also enabled the researcher to make rational engineering judgments of the problem. In conclusion, primary data collection is ideal given a large budget and adequate time frame. On the other hand, an accurately recorded and reconciled historical data can be used for analysis of general cargo port with a high confidence level. In the author's opinion the results achieved by using accurate historical data will not be significantly different from a similar analysis using primary data. 4.2 Design of the Data Collection Forms The development of detailed collection forms is a prerequisite for an in depth analysis. Data collection forms should be tailored to accommodate information concerning the variables discussed in chapter 3. The time ships and cargo spend at different port subsys- tems and other service variables necessary for logistical coordination need to be recorded. It is important that data forms should be de- signed to recognize the sequence of events which occur as ship and cargo move through the port system. 105 Tables (4-1) through (4-7) illustrate the data forms developed and used in this study. Tables (4-1) through (4-6) are aimed at identi- fying ship and cargo delay at different port subsystems while table (4-7) yields the capital and operating costs of specific port investments. 4.3 Port Mornhology . Lagos port complex has a unique geographical advantage. As shown in the location map, it lies on the lake 2.5 nautical miles from the Atlantic shore lines. This means that it is sheltered from high Atlantic waves. Additional protection is achieved by three man-made moles. These moles act as breakwaters (1) and have been effective in the protection of the harbor. The major problem with the moles is rapid settlement. The cost of restoration and maintenance has been a burden to the National port authority. On the average the depth of water in the harbor is 9.14 meters which can safely buoy large ships and ocean tankers. Figure (4.3-1) illustrates the geographical location of Lagos port. The quays cluster around Apapa which is a mainland suburb of Lagos Island. In administrative terms these quays have been organized as three distinct ports: (2) - Apapa port 23 berths - 3rd Wharf Extension port 6 berths - Tin Can Island port 10 berths Total 39 From the point of view of traffic and logistics these 'ports' are indeed one complex port system, because the controller of Harbors directs ship movements in and out of any of the above ports, and 106 .. .. . fin .......:. I. .1 Mae? anon: unm anmanxu . s . ... nEa oaam .2 NE non NM ouonm zaaaauon ao aonEE< oaooano> aoaa ou . o onoam . . aoaoa aooEE< aoEEE< ooa>nom amou aoaanou amou aoaanou oaooaao>< anaooom momzm hamzza 4<:zz< xmanzou amen mow<4 .Aaiv.¢v manoa 110 oo.NNm.mmN.mw n on.eom.aan.a a oonoméfla a. anEamo>Ea ao Eanoz anmonn aoa ao amonana Eo mEazoaa< aoaoa oEono aonEE< anmanxm oom.mom ooo.mmm oom.ooa .mnz EN ooo.omm.m oo.oNN ooo.ma anoEz onm anon oanma OON.mam oom.aan oom.nmN .mna ON ooo.oen.a oo.oNN coo.aN Eou Eaa ozone cow.nmo OQN.mNm coo.Noa .mnx EN coo.NmN.m oo.ama ooo.ma anmoo anon ooa.moN.a oom.mom oom.Nco .mnz oN ooo.mmo.w oo.ama oom.¢e onon< mo mo EoE mo>E E ooon new: 3% .. 1.. a... snow... mmmzozmmza m<=zz< xmmnzou anon mow aaaanao onaa auwwmmw eeaaaanmwwm aoEEE< ao3EE< oua>nom . czawamzo uzzooxza 4<=zz< xuanzoo axon mooEa E moonm aonEE< oaooano> aoaanou oowwwwm amoo muaanou amoUNfiouwmou waonoam zawwmmmwoflo aoaoa ao=EE< aooEE< . . . . oaooaao>< . mu2a 4<=zz< xmmnzou axon mowEa no Eanoz aoomonn Rea no amonmaEa Eo mEazoaa< oo.¢om.nmm.m a oo.oon.o¢m.¢.z 'Iflfl aoaoa oEono ao=EE< anon . oanma oom.omm.a con.nom.a .mnz om cem.mmm.no coo.nN oom.N Eou Eaa onon< anmanxm oom.oao.a OON.NoN con.wom .mnz om ooa.omo.o¢ coo.NN oom.a anon: onm mzooo oom.nn cea.am cem.oa .mnz om m.oom.on m.oma.N mam anmau anon ooo.ooo.a ooo.oam Noo.amo.a .ann om ooo.ooo.am. n.~om.ao mom.N oooo< .z. amou .z amou aoaanou oaa._ t .z noaoE Eaman eaaaaauoa ao aoEEE< aoaoa oaooano> ao=EE< ooa>nom amou aoaanou non amou zone anaouom mzazmm Hzzooxfinm amou azmzamm>za a<=zz< xmmnzou anon mow aoaanou oaam maaEz anananommn aeooo< aoaoa noooo< nooee< ooasnom .oaaooo aaon ana asznazou oz< az<4n wzaaozza a<=zz< xmanzou axon mowEou m oom.mNa ona.ao onN.No ca ono.NNo ooo.nma o mooeoa mm-m~ noo.mm ooN.aa ooo.NN ca noo.oNN ooo.om o mooooa NN non.ma oom.o Nma.o ca on.am amo.mm o mooooa no anoaaaa aooaonn n oom.moN non.no oom.mMa ca noo.mom.a noo.mm no noaaannoaoo o oom.ono ooa.oma on~.oam ca noo.~oa.m ooo.om mo onoaoona m mna.nao mNn.moN omo.aao ca oom.oaa.o moa.ao sea onoona anaannon omo.ooN ooo.ao moo.Noa ca omm.oNo.a aNo.NN on ooeooa o mnn.mom omN.amH mam.NoN on oma.mNo.N noo.oa Non mooooa m non.moN noo.ao ono.moa na ooo.omo.a noo.oa an mooooa «-mm manaannon o .z amou .z amou mnooa . a. as a o o E. anon z 38 o oz aooee< aoaoa oflmowno> ”muhomo oowwwwm aoaaooo one: oaaeo ooaa anoooo ann osm.oom.e unoznoa azmzonm ooooaaeoo An-o.ov oaooa oom.~oo.oa u noaoa oom oma.ooo.m n noooa moo.oa ono.o omn.o ca oom.no oom.oa m nooooa aaom 5. man m moonw nun oon.aN oom.n ooo.oa on ooo.ooa omN.oa o ooaoEoo omo.o oaN.m oNo.o on ooN.oo omo.oa o mnooa>oam noom moo.oa moo.o ooo.Na ca ooo.oNa ooo.Na on anonooam oom on ooo.oom.m ooo.oNa.a ooo.omN.N ca ooo.ooo.mNN ooo.oNa ooa onoaaona o .z. amou .z amou mnooa a. 38 o a. :8 2 as nooano> aoaaooo onao mane: eoaaoanooon Eoon aoEEE< aoaoa aooEE< aanE< oua>nom aoaanou aaE: NEN.oaN.n onoznon aoooonm oooeaoooo .nn-o.ov oaooa 117 Noo.ooo.oa a u aoaoa oom Nmm.mNo.m » u aoaoa ooo.nm Nno.ma own.aN on onN.oaN ooN.o mm “wwwwwuwo no ooa.oN ooo.o. oma.oa on ooN._oa ooo.o mo awumwwmm“n on ooo.oo ooo.oa ooo.o~ no moo.oo~ amo.NN ma mnoanaa aooaonn ma Nom.no ama.oN ao~.oo on ono.Nom mo.omn.oo o ammommmhn on ooo.om ooN.Ma ooo.oN on ooo.oo~ ooo.o oo moaoom ooaooaoz ma ooo.ono ooo.oo~ ooo.oom on ooo.ooo.m ooo.oo~ om onoo oona aonn ooo.noo ooo.ooN ooo.oom no ooo.ooo.m ooo.omN om onao onna xom mEomoz Eozaaon Na ooo.moo ooo.mam ooo.omo on noo.oom.o ooo.ooo n on com. ooo.ooo ooo.ooa ooo.oom on ooo.oo~.m ooo.ooo o no moo. mo>aaoEoooa aa ooaaooom .z amou z, amou mnooz . a. :8 o a. 3.8 2 :8 oz noaanon aooanoo onaa mono: ooaananooon aoooeo aoaoa noooo< aaooo< ooasnom aaaaooo one: ann oom.Noo.oa onoznon aeooono .zo-o.ov oaooa 118 85.8%.: n aoaoa oom ONm.nNO.o n :38. . . . . . mnozo>Eou OOO Na OOO o OOO O Oa OOO om OOO Oz o 228 ON Omo. we no: . 2 8m . Nm Oa OOO. mNm OOO. 2 2 no 23.. 2 . . . . . . onoaaona NNN mom non noa mam mom Oa OOH omn m ONo oma nN zoonaz ma .2. amou .z 3.8 mnooz - l . .z anon o z. 38 .z 38 oz aooano> aoaaooo onao oaaon ooaaoanooon aooeez aoaoa .oooo< aooooo ooasnom aoaaooo aaon Ewan macammoiflnw u ULQEOL. H£OSOLm tweawucou .AmleVV mwnmh 119 oo.oom.NOo.ma am: aoaoa cam oo.ooo.Noo.naa u aooEamoson no zanoz aoooona no.0oa.aoo n .oaoa oom amonoaon aoa ooazoaao OOa.mN NNn.n Onm.ma Oa aON.mma mmm.¢ mm momam oEoz mN moa.oOn OOm.Om ONN.NN OH OON.NNN ON0.00 O mnooooa ao>ozm NN . . . . . anEnaocm OOO ma com a OOO O Oa OOO OO OOO O OH OEo ozoom aN .z amou .z amoO mnooz .z amoO o .z amou .2 amoo .oz aooano> aoaanoO oaaz maaEO anananumoO aoOEE< aoaoa aoEEE< aoOEE< ooa>nom aoaanou aaEO ana OON.HO0.0H onoznoa azmaonm .Amie.ov oaooh oo~.~ce.maunnm= aoaOP oo.oom.mHm.ean u anEamm>EH ao canoz aEmmmnn 120 c.8536 n z. aoaoa :2an E: oenzozz OOa.am0.0 Omn.mmN OOm.n~m Oa OOO.mnH.m OOO.mNN ON mEoanon O OOm.OOO OOm.mOa OOO.aOm Oa OO0.0aO.m OOO.maa om momnom O OH OOm.mOn OOa.mON mam.Omm Oa ONa.mOm.m mmO.NOo ma mmna noono: o mN.ma oOoE -an onoaaoO maocz Na OOm.Noa OOm.no OOO.mO OO OO0.0mO OO0.000 a n; OOo.a. 3onom Eaza. m maocx m.aa Oma.oN Om0.0 OOO.aOa Oa OO0.0aO.a OOO.mOO N n; OON.a 3onum Eaza N mm.NN oOoE ian onoaaom maoEx Na OO0.000 OO0.0Na OO0.0oN Oa OO0.00¢.N OO0.00N.a N .nz Ono.a. zonom Eaza. a .2 amou z. amoO mnoo» .2 amoO .2 amou .z. aaEO non anananOmoO aoEEE< aoaoa aoOEE< aoOEE< ooa>nom mmozza 4<=zz< xmmnzou axon moo_¢¢< nazm 1} ...-o... onooan ADNHOOBUJ aca rON 123 ' Hence the definition of arrival and service distribution is a pre- requisite for the design of a simulation or an analytic model. The initial step in the fitting of any distribution to traffic data set is to assume a know distribution (i.e. Poisson, Binomial, negative Binomial, negative exponential, Pearson type III, Erlang). For this study a Poisson arrival rate was assumed while service rates was assumed to be negative exponential. The observed distribution is then analyzed to determine whether the postulated distribution is the true population. The Chi-square test is widely accepted as an impor- tant index of the goodness of fit of observed and theoretical frequencies of a data set. (10) Chi-square technique is summarized by the equation: x2 = Z (f-F)2/F where f = observed frequency F = theoretical frequency (f-F) = deviation of any cell. Finally the Chi-square value computed is compared with values ob- tained in Chi-square tables to determine the probability that such values occur by chance. Figure (4.5-1) illustrates the ship arrival data frequencies. using a time interval of one hour. In table (4.5-2) the null hypothesis postulates that arrival distribution is Poisson, i.e. _ e-mux . . . . (11) P(x) x! Where x indicate random traffic events. u = mean number of occurrences of these events. 124 Table (4.5-1). ARRIVAL DISTRIBUTION (Inter Arrival Times) Seggal Name of Ship Date Aiflivgi Arrival iiis) Remarks 1 George Armfield 1-7-78 0015 0 2 East Wind 1-7-78 0615 6 3 Joselin 1-7-78 0630 .25 4 YATSENYAVI 1-7-78 1225 5.92 5 PLAYA MAS 1-7-78 1600 3.28 6 NORDRAP 1-7-78 1615 .25 7 PLAYA BLAMCA 1-7-78 1830 2.25 8 MOURA 1-7-78 1835 .08 9 VASTIRAM 1-7-78 1930 .92 10 STOIK SUND 1-7-78 2255 3.42 11 ANNA WESCH 1-7-78 2345 .83 12 FELIPES 2-7-78 0800 8.25 13 MEVENBURG 2-7-78 0910 1.17 14 GREAT MAURICE 2-7-78 1050 1.67 15 AGAPPI (SWLA) 2-7-78 1400 3.17 16 MESSIMIAKI 2-7-78 1730 3.5 17 BABOUNIS COSTAS 2-7-78 2200 4.5 18 FRIO DOLPHIN 3-7-78 0500 7.0 19 TINITO CASTRO 3-7-78 0700 2.0 20 PIRAN 3-7-78 1245 5.75 21 LIDNIGER JADE 3-7-78 1345 1.0 22 BRITISH WILLOW 3-7-78 1615 2.30 23 DANAFRIO 3-7-78 1910 2.88 125 Table (4.5-1). Continued Sfigial Name of Ship Date Algivgl Arrival (gig) Remarks 24 Jolly Azurro 4-7-78 0400 1 8.17 25 SIMONA 4-7-78 1000 1 6.0 26 DOLLIARI 4-7-78 1400 1 1.0 27 BRITISH TENT 4-7-78 1405 1 .08 28 ESPRESS SARDEONA 5-7-78 0810 1 18.08 29 BLUE AKEISHI 5-7-78 1035 1 2.42 30 MATADI PALM 5-7-78 1200 1 1.42 31 PETRELO 5-7-78 1445 1 2.75 32 ESPRESSO SICILIA 5-7-78 1710 1 2.58 33 THOMAS WEHR 5-7-78 1800 1 1.83 34 WEST 5-7-78 1810 1 .17 35 THUMTANKI 5-7-78 20,00 1 1.83 36 APAPA PALM 6-7-78 0830 1 12.5 37 BOREA 6-7-78 1200 1 3.4 38 ORMOS 6-7-78 1600 1 4 39 STEFANIA 6-7-78 1830 1 2.5 40 ENA SIF 7-7-78 0625 1 11.5 41 FRISIAN TRADER 7-7-78 0800 1 1.75 42 RHOMBUS 7-7-78 0900 1 1 43 NELKAR 7-7-78 1000 1 1 44 DORT SKOU 7-7-78 1100 1 1 45 STINNES ZEPHIR 7-7-78 1436 1 ‘ 1.6 Table (4.5-1). Continued 126 Serial Time of Gap No. Name of Ship Date Arrival Arrival (hrs) Remarks 46 ATALANTI 7-7-78 1605 1 1.52 47 PEP SPICA 7-7-78 1848 1 2.95 48 JANNE FREM 7-7-78 20,00 1 1.8 49 LEILA BECH 8-7-78 0600 1 10 50 METTE BRAVO 8-7-78 0715 1 1.25 51 SOL NEPTUNE 8-7-78 0750 1 .75 52 TINE BECH 8-7-78 1030 1 2.67 53 PANDA STAR 8-7-78 1155 1 1.42 54 PAOLA MONTARI 8-7-78 1220 1 1.27 55 CLAUDIA MARIA 8-7-78 1336 1 5.67 56 ALRAZAK 8-7-78 1910 1 57 ALBERINO 9-7-78 0400 1 7.83 58 YUE FLOWER 9-7-78 0545 1 1.75 59 INLAND 9-7-78 0600 1 .25 60 FREEZER FINN 9-7-78 0730 1 1.5 61 FLORA 'C' 9-7-78 0845 1 1.25 62 FRISIAN STAR 9-7-78 1145 1 3.0 63 BRITISH LOYALTY 9-7-78 1315 1 1.5 64 BRITISH TRENT 9-7-78 1400 1 0.75 65 PARTULA 9-7-78 1416 1 .27 66 ODETTE 9-7-78 1630 1 2.23 67 SIMOMA 9-7-78 1900 1 2.5 127 Table (4.5-1). Continued Sfigial Name of Ship Date Altivgl Arrival ?:ES) Remarks 68 ALCARA 10-7-78 0030 1 5.5 69 AVRA 10-7-78 0230 1 2.0 70 DORA BALTEA 10-7-78 0345 1 1.25 71 SASSANDRA 10-7-78 0930 1 5.75 72 WOERMANN 10-7-78 1530 17 6.0 73 LIBRA 10-7-78 1530 15 O 74 EXPRESSO 10-7-78 1630 1 1.0 NB TWO PIEMONTE arrivals 75 LASS '10-7-78 20,00 1 3.5 128 m = mean of the Poisson distribution which is also equal to the variance. As shown in figure (4.5-2) ship service times were also classified into frequencies. In this case the null hypothesis postulates that service times follow a negative exponential distribution curve. In table (4.5-4) the Chi-square statistics was applied to the data set to determine the validity of the initial hypothesis. The probability density function which corresponds to the nega— tive exponential distribution is given below: (12) f(t) = qe'qt where q is the mean flow rate in ships for unit time. t = unit time interval. It is important to note that the average service time (t) varies inversly with the flow rate. In both tables (4.5-2) and (4.5-4) the basic assumption is that the probability of a headway between t1 and t2 is equal to the proba- bility that the first event is between t and t2. (13) 1 P(t1 < h < t2) = e'qtl - e‘qtz As shown in figure (4.5-2) the mean arrival frequency of 2.89 hours per ship indicate that .35 ship arrives in the port of Lagos every hour. The probability of ship arrival times occurring within an interval of half an hour is .32. The cumulative frequency (f) between an interval of 1-2 hours indicates a good fit with the theoretical frequency (F). In general the results of the Chi- square test indicate that ship arrival distribution follows a Poisson distribution as postulated earlier. 129 Table (4.5-2). CHI SQUARE TEST APPLIED T0 ARRIVAL DATA was“ ..Szizzzidm Pm 1.2223235“. .... an... 0.1 21 .32 24 18.4 1-2 20 .27 20 20.0 2-3 11 .20 15 8.07 3-4 7 .12 9 5.44 4-5 2 .06 4.5 .89 5-6 6 .02 1.5 24.0 6-7 1 .00 .00 7-8 2 .00 8-9 1 .00 9-10 1 .00 10-11 0 .00 11-12 1 .00 12-13 1 .00 13-14 0 .00 14-15 0 .00 15-16 0 .00 16-17 0 .00 17-18 0 .00 18-19 1 .00 .15 76.8 Mean -- 2.89 x2 = ‘2— fZ/F - R Variance = 2.89 3. X2 = ;gjé _ 75 = 1.8 Accept Poisson x0.05 = 9.488 > 1.8 distribution. Hence arrival rate = l_= _____= .35 ship~ per hour u 2.89 130 it com ONa . coma Outsiaa coca astute zaowmz ma VOa OOH.¢ 00a omnH Omimim ONma mutate z<>oz ma OOO mam.O ma omaa mummim coma maiNia au Ma c _it an coca mansia OOHH OmiOimN N a nan sagas ”one .....o an“... ...; ..EM. e .2 zoaaamamamao mo~>mmm nazm .nm-m.on oaoon 131 wmm ONe.m Om OOno muumiOa OMOH mutate“ onocommom mm ii moe.m ONO Omno ONiNiON omOH Outsiea nmucomaon nm OOa.N ii mOm coma OutsimN coca Outsima azncoama no zaau Om ii OOO.H oO cana whiniea coma ONiNiNH anamaox< oaam Om ii OOO On Omma Outsiea mesa ONINiaa moazunmz om ii OOO oHa omno ONINiOH coma Outsiaa noaooz ouma> mm OOO OOO.N Nu OOOH OmiNiOa OOOH Ominuaa nmzoan oO> Nm ii «BO Oaa ONHH mmimima OvNa ONTNiOa onoan am ii ON¢.H maa OOOH whiuima OOOH ONINiOH Oz ON swam ...wmmufia flaws 2.: gamma 2.: gamma as no as .oz ooooaaeoo .nm-o.ov onoaa 132 in Nem.~ ONO Oama ONiOiO OOON ONiNiNa >ozm NO in nww.m Omm OOO“ whimuO ana ONINTON OOn NO ONH mON.N OO OOOO ONiOiO OmOH ONiOiH Oszzamoz OO OO OOO.m OO OOHH ONiOiO OmOO ONiOiH mO¢mm nazm an .aNumv onEOan AONBROSHJ ‘— Table (4.5-4). 135 CHI SQUARE TEST APPLIED TO SHIP SERVICE TIME AT BERTHS 121:3“ #23322; (n R» .Izzizizicil, RZ/R Ramses 0-100 26 .39 29 23.3 100-200 20 .24 18 22.22 200-300 10 .15 12 8.33 300-400 10 .08 6 16.67 400-500 5 .06 4.5 5.56 500-600 3 .03 2.5 3.6 600-700 0 .02 1.5 0 700-800 1 .02 1.5 .67 75 75 80.35 Mean = 199.00 (hrs). x8.05 = I fZ/F _ R = 8.31 days. XS.05 = 80.35 - 75 = 5.35 From Chi Square table x3.05 = 12.592 > 5.35 Hence negative exponential distribution thesis is correct. Service rate = is accepted, i.e. null hypo- = .005 ship per hour per berth. 3. Total Service rate = uK = (.005)(39) = .195 ship per hour where K = total number of berths. 136 Table (4.5-4) shows that mean service time for ships in the port is 199 hours. This means that .005 ship is served every hour per berth. When the entire 39 berths are considered the average service rate is .195 ship per hour. Hence the berthing capacity is far below the arrival rate of .35 ship per hour. This reason explains the in- finite nature of the ship queue in the port of Lagos. In addition table (4.5-4) indicates that the probability of ship service times between 0-100 hours is .39 while between 100-200 hours the probability drops to .24. In general the service distribution fits a negatiVe ex- ponential curve as proposed by the null hypothesis. 4.6 Analysis of Cango Deiey Analysis of cargo delay within the port subsystems is important for the following reasons: - Cargo delay within transit warehouses provide the basis for deter- mination of warehousing cost in ton hours. - The rate of cargo flow through warehouses is a logistical criteria for determination of the size and number of warehouses required to accommodate a given daily tonnage. - The type, volume and class of cargo moved by direct and in- direct channels are the basis for determination of the level of improvements required in various port logistical subsystems (i.e. warehouses, rail cars, trucks and handling equipment). - The distribution of cargo waiting time yields average time for determination of associated costs of handling, insurance and storage. Cargo loaded or unloaded from general cargo ships moves through two major logistical channels. 137 ° Direct (into trucks, rail cars) - Indirect (into transit warehouse and open storage areas). Data obtained from the Lagos port complex indicates that in 1978 cargo moved via direct intermodal transfer made up about 83%(14) of total ton- nage handled while indirect cargo tonnage was 17%. These figures are significant as there is a long cargo waiting time (109 hours) for in- direct cargo movements. These delays create a tremendous congestion problem in both the transit warehouses and the open storage areas. Estimation of Direct and Indirect Cargo Tonnage: Million tonne Total Annual Through Put (1978) = 8.99 Total Liquid Cargo = 2.43 Total Dry Cargo = §;§§_ Direct Cargo = .83 x 6.56 5.44 Indirect Cargo = .17 x 6.56 1.12 Distribution of Cargo Waiting Time: As illustrated in tables (4.7-2) through table (4.7-5) a sample of 56 units of cargo was examined. Note that only indirect cargos were analyzed because there was no waiting time associated with direct cargo. Both the weight and the total time Spent by each unit of cargo was determined. Then the average time spent by one ton of cargo was calculated to determine the annual cumula- tive time spent by indirect cargo during the clearing and forwarding process: 1.12 Million tonnes Total Indirect Cargo tonnage (1978) Average waiting time per unit 109 hours Cumulative Cargo waiting time 109 x 1.12 million ton-hours for 1978 122,090,000 ton-hours --(a) 138 Breakdown of Transit Inventory Cost: Inventory related cost can be summarized as follows (8) Total Costs = Ct + Cs + Cc + Cin + Cob + Cord where Ct = cost due to tax C5 = cost due to storage Cc = cost due to capital Cin = cost due to insurance Cob = cost due to obsolescence Cord = cost due to order processing and handling. In this study only items which constitute costs to the port will be considered. In addition the port of Lagos is a public port and is exempt from tax. Based on the data obtained from cargo supervisors and traffic officers the cost table below was prepared. $ ¢ Ct per ton/hr = -- -- Cs per ton/hr = -- .25 Cc per ton/hr = -- .25 Cin per ton/hr = -- .50 Cob per ton/hr = -- .25 Cord per ton/hr = 1 -- Total Transit Inventory cost per ton-hour = $2.25 ----- (b) The total cost of the annual cargo waiting time can be obtained by multiplying items (a) and (b). i.e. $122,080,000 x 2.25 = $274,680,000 per year. The above cargo delay analysis indicate that there is a great need to reduce the average cargo waiting time. The causes of this delay will be identified by the logistical operations survey. 139 OONO OOOO N O. ii O.¢O ON\OH\HH ON\OH\HH .. OH O0.0H OOOH N N. i- 0.00 ON\NH\HH ON\O\HH .. O OOoH OONH N o. 1. OO ON\NH\HH ON\O\HH .. O OONO ONOO N O. .. O.HOH ONxHH\HH ON\N\HH .. N oooH omoH N O. .. O.HO ON\O\HH ON\O\HH i- O OOOH OONO N N. .. m.ONH ON\NH\HH ON\O\HH .. O OOOO OOOO N O. .. N.HN ON\N\HH ON\¢\HH .. a l OONH OOOH N N. .. N.OHH ON\O\HH ON\O\HH u- m OONH OOoH N O. .. N.¢N ON\O\HH ON\N\HH i- N OmNH OOOH zoan u N OH. .1 o.HN ON\¢\HH ON\H\HH .. H ooanHO om>aooom Hmnzv H.mnEV anEoEm oaoO maoO oEonO znanO .monza HmEONv moaaaaaoon Oonoa Ammoonm Eono noOEO oEaN -nonoO Haom OEo aEanz nonmEonN HoEnoa Ea oEaN OEo omzozonozv OEaaaoz OOnoO .oz ozoonNiimxnoEon -EH Ea oEaN moaaaaaoon oOonoam EanoO anH AOZOHNOzOO mzHN «On mamNmzaoooz amazv .monoa mnzoz maoO oaoO, mEonO nanO inonoO Haoa OEo Hmmwwv no mewaaawwwmn MOonoaw Amouonm Eono noOEO oEaN szEnNiimxnoEon aE . 3 a i: NEH oE a Ea oEaN OEo omzozmnozv OEaaaoz OOnoO .oz _ a an moaanaaoon ooonoam zanom Eoon AszHNOZOn mzHN «On OOONNZ<¢aoomm Hmnzv monoa HmnEv o -nonoO Haom OEo amEoNv moaaaaaoon Amnzozv oaoO oaoO oEonO OnanO mxoonNiimznoon aEanz nonmEonN HoEnoa oOonoaO mooonm Eono noOEO oEaN -E Ea mEa Ea oEaN OEo omooEonozv OEaaaoz,oOnoO .oz n . .n moaanaaoon ooonoam zanoo EooH nmzooaozzn mznn Eon mznnnzaouon HmnEv . Amnnocv oaoO oaoO oEonO NEanO an .. a... m onnOiww mono a; . 3 a -E NE” oENa Ea oEaN OEo mmnozonozv OEaaaoz OOnoO .oz n N n n n . .N monaaaaoon ononoam nonon ann AmonNOzan mzHN zen OOONOz< N.Ona n mn< nnoH nnNH n O.N -- n.nnN nNnanH nN\nN\NN -- no ,i -,. -- .: .mnHH nnNn N n.o n- N.mnH nN\n\a oN\nN\NH -- on noHH ann N n.N -- n.noN nN\n\H nnnNnnnH -- on nNNH nnHH N N.H -- n.oo~ nNnnnH nN\NN\NH -- no nnoH nnoH N o.N -- H.Non nN\N\a nN\NN\NH -- on nnma nHNn N n.n -- N.OnH nNnnonnH nN\HN\NH -- Hn A92: Nam—5.2.0 wm>qumm AWOLNS mnao 0 O m mew-a n :O nan one .. swoon”. Mesa - {sane ..n .w onnOiww mono aE . 3 a in NE” oENa Ea oEaN OEo momnozonozv OEaaaoz OOnoO .oz n N n n N . .N moaaaaaoon ononoan zonon ann nonnaanon .zn-n.on oNnoN 10. 11. 12. 13. 14. REFERENCES Handbook of the Nigerian Ports Authority, Development Department NPA, Lagos. Special Tin Can Island Port News, Nigerian Ports Authority, Lagos, £0me Nigeria. ' Ibid. Ibid. Ibid. Tin Can Island Magazine: Nigerian Port Authority, Lagos. 197 Ibid. Ibid. Ibid. Drew, Donald R.: Traffic Flow Theory and Control. McGraw-Hill, New York. Ibid. Wohl and Martin: Traffic Systems Analysis for Engineers and Planners. McGraw-Hill, New York. Ibid. Bowersox, D. J.: Logistical Management. Macmillan and 00., London, 1974. 145 CHAPTER V SHIP QUEUING SIMULATION MODEL 5.1 Assumptions The following assumptions were considered in developing the simu- lation model: - A first come first serve queue discipline exists at the study port. - All berths have the same service time (i.e. there is no significant variation in berth service time). - There is no reneging or any distortion of the queue discipline. - Any of the berths can service any class of ship entering the port. - Ship arrivals into the port have a Poisson distribution - Ship service time follow a negative exponential distribution. In the first place, the port of Lagos has an established first come first serve queue discipline. However, when any cargo or ship is considered a priority class, floating cranes are employed in un- loading the ship. This means that the established queue discipline is not distorted. The second assumption can be justified because all berths utilize the same number and type of cranes, fork lifts, and warehouses. These equipments are drawn from a common pool. In ad- dition labor gangs assigned to each berth are the same. Thirdly the strict Harbor master's control policy and towage services prevent any reneging. Hence there is no distortion to the established queue dis- cipline. As discussed in section 4.3, the minimum depth along the quayside is 9.14 meters while the maximum draught of ships calling 146 147 in the port of Lagos is 8 meters. This means that any berth can pro- vide adequate draught required to buoy any ship. 5.2 Logic Diagram and Model Variables The logic diagram is illustrated by figure 5.2-1. The program is written in a simple fortran language using the University of Minnesota fortran language compiler. The model starts with initial- ization of the model variables: RSHIP - Ship arrival time R Service - Ship service time TRSHIP - Total ship arrival time TR Serv - Total ship service time ATRS - Average ship arrival time AR Serv - Average ship berthing rate In the initialization process it is important to specify that arrival rates and service rates have real value. This specification eliminates negative arrivals and service which are absurd in this situation. Secondly, the number of ships are specified as integers for the same reason as above. R Ship and R Serv are time dimensions during which ship arrivals and ship service events occur. Finally the initialization process indicates to the computer the number of observations required. The next step in model building is to specify ship arrival and service rates. Ship arrival rates is the total arrivals at the study port in one hour. Also ship service rate is the total ship service offered by all the service berths (i.e. number of berths, average berthing time). Definition of ship arrival and service distribution 148 is a major step in the simulation model. Integral transforms are powerful tools for simulating ship arrival and service distribution. With the input above the computer generates and prints ship arrival and service events (i.e. random numbers). Do loop 500 in the logic diagram is responsible for generating these random numbers. The Do loop 300 prints the array of the random numbers generated. The next major step is to define the variables (TRSHIP, TR Serv, RSHIP, ATRS, AR Serv). These variables are defined with respect to a time axis. At this reference axis time is set at zero and all 39 berth facilites are regarded occupied by ships. Hence the time of t“ “ arrival plus the sum of the (i-1) arrival of the i ship = the it arrival times. Ship service time is also measured in the same manner. The average shipberthing rate is set equal to TR Serv/P where P is the number of ships which have been serviced since time to. The queue is defined by 00 loop 465. A queue exists when TR Serv is less than TRSHIP. The logic is that the arriving ship cannot find a vacant berth and is forced to wait in the queue. When a vacancy occurs in the berths, the vehicle at the head of the queue moves to the vacant berth. The model updates the ship arrivals and service using the list of random numbers generated in do loop 500. Additional ships are added in the queue and the simulation continues until the specified ship number is reached. A list of the computer cards and model format are shown in figure (5.2-2). 149 Figure (5.2-1). SIMULATION LOGIC DIAGRAM START [INITIALIZATIONJ SPECIFY SHIP ARRIVAL AND SERVICE RATES DEFINE ARRIVAL AND SERVICE DISTRIBUTIONS GENERATE RANDOM , NUMBERS I DEFINE ALL VARIABLES I f Total Ship Service Time is Less Than TotalArrival Time -Yes ==4 BUILD A QUEUEI UPDATE ALL SYSTEMS ‘Tfi Is Simulation Yes Complete? =:% Print Resultsj END 150 n\znO>nnmnNuznnnnmnN< nnnNNnn .oN zH-av>nnmnN+nnO>nonnuznn>nnnzN nnNNNnn .nN HH-nOnnzmnN+navn_zmnunavnnzmnN nonnNnn .oN H.nnm.Nun nno nn nonnnnn .nN zHO>nomnNnnaOnomnN< anmNnn .NN nHO>zomnquO>nnmnN nonmNnn .HN annnzmnqunnnznnN nNNNNnn .nN nnanznn nnn nonNnn .nH non.N.anJ»n.NnanONnomn.znnnnzmnnnno.nNOmNnnzn.nunn.nnnn nH nnnNNnn .Na H.na.Hun nnm nn nnNNNnn .oH mnanznn nnn nNNNNnn .nn non.N.-an.Nn.N-.manannnn no nNNNNnn .oa a.Annnnnmn.nannnnnnnnn.nNOnNnnzn.n-.no.nnnn nH nnoNNnn .nH oNnnnm\ x-H nno<-uz_ nnomn nNNNNnn .NH oNnnnn\ N-H nno<-nnn nnzmn nnHNNnn .HH xnnznnnx nNHNNnn .nH Nnnznnnn nHHNNnn .n H.nnm.ann nnn nn nonNNnn .n non.unNv OOOHnZOO zNHOOO>HZO .OOOOZ zOHNnnnnN.znvnnzmnNnnno.nNOmNanznn.nnnn.nnnv no H.nH.Hna.noo nn . nnanznn OOO HOO ONHH OOHH OOHH OOO Nme OOO ooOEaanu OHOONOO OOmONOO OOOONOO OOOONOO OHNONOO OOOONOO HOeNOO OOOONOO OOOONOO ONOONOO ONmeNOO ONNONOO OONONOO OHNONOO OHNONOO OONmNOO OOOONOO ONOONOO new .AN-N.nO anonnn 152 5.3 Sensitivity of Total Shig_0eley to Increase in Number of Berths The simulation model was utilized to test the sensitivity of ad- ditional berths on ship delay. In this test the service time is held constant and equal for all berths. The program creates five new berths each iteration. This means that the port service rate SERATE is in- creased each time by 5n (where u = .005 ships/hr. = average berth service time). With each new service rate the port was simulated and new ship delay and queue lengths determined. The number of berths was constrained such that the utilization factor would be less than unity, i.e. (1) o < 1 where p = A/X8u A= ship arrival rate per hour X8: number of berths u\= average service rate Table (5.3-1) summarizes the reduction in ship delay due to ad- ditional berths. The delay parameters in table (5.3-1) were obtained by considering the 100th ship arrival. As shown in table (5.3-1) total ship delay is very sensitive to additional berths. The addition of 20 new berths reduced delay be 76% while the addition of 25 berths reduced delay by 87%. A diminishing return is observed with additions of 30, 35, and 40 new berths. It is also important to note that with additional berths the que length decreases. In figure (5.3-1) the queue wait- ing time associated with a corresponding number of berths was plotted. This curve provides a basis for evaluating whether the reduction in 153 mar—Run mm «was: onuaa OO.ON ma.aN nN.nn . no on NO on 'I'I'O'" . I . .. d 1 i 1 o ."” II .a. III] II: .cm To .. .222 r/ I II Jail m [I 38~H I V. I I . ,9 n“ 9 m J]! I. II .... , .nnamw .35. non .2: OH L ’1 m oz: 3.5% a 8 55.. 9:55 ...I i .. ...... .r m I ad I S I z. . , nno o o t a .n H ..noo .onn nnn u we .nnn .nnzzzon nNnH .o.aa nznnNnnzno non>nnn ”i=2: SE: 2:: no nonznz zn Hafiz. 2 oz: 2:2: one no NN::N_mznn inn. 23: 154 .Amumwxm msmsc a cmzzv “Lon mg» sag; mawgm mo mum; ogaugmawu op mucoammggoo mpsh .Lzuoo mama mo_>gmm saws: am can; mgu pan gugmn um mum; mow>gmm mgp no: mm mum; m:_;ugmm. .ugoa xuzum msu cw mgugma mo Luggac m=_pmwxm mgu my mm« mmH co“ o o.ooH o .mg; mm.~ om mm“ co“ 0 o.ooH o .mg; ~¢.N mm mm“ ooH .o o.ooH 0 .mg; om.~ om New mm m o.mm, m .m;; em.~ ox com um N o.mm o .mg; mm.~ we mum mm on m.qm mH .mg; NN.m we mmm mm Rm N.co mm .mg; m¢.m mm mmw No om m.mm mm .mg; Hm.m em mum me mNH m.¢m mm .mg; ow.¢ me ohm mm as“ Mmefi NR .mg; wo.¢ cc one mmm Q_;m mco mung Amggv use; c? Amgsv mamas :w Ama_;mv acwcugmm mcugmm ampoo Page» cowuuzumm a mswp mcwp_az corpozuwm & sumac; mac mango>< Lo Logszz mzhmmm no mmmzzz z~ hszmmqu OH ><4mo no >h~>mhmmzmm .Amum.mv mPQMH 155 delay due to an incremental increase in the number of berths is linear. The curve in figure (5.3-1) indicates that the slope of the curves with- in a given range of berths approximates a straight line. Hence a gen- eralized equation can be developed to relate the queue waiting time (X1) and number of berths (X8), i.e. X1 3_A1 - A2(X8 - M) where A1 is the queue waiting time associated with M berths A2 is the slope of the curve X1 is the queue waiting time X8 is the number of berths required for a queue time of less than 3 hours M = a specific number of berths. Figure (5.3-1) indicates that when the number of berths is between 64 and 70 the slope of the curve is approximately 4.4. In the range of 49-54 and 54-59 berths the slope of the curve approximates to 7.8 and 6.6 respectively. It is also observed that between 64 and 70 berths the queue waiting time is tolerable (ranging between 30 hours -- 2.85 hours). Hence the optimization constraint for the linear programming problem should be derived in this range, i.e. x _>_ 30 - 4.4(x8 - 64) 1 The simulation results illustrate the sensitivity of the queue waiting time and queue length to incremental increase in the number of berths. The addition of berths to the port of Lagos is one of several actions which could be taken to reduce delay and associated cost in the port of Lagos. Additional units of equipment, warehouses and labor gangs might reduce ship berthing time and subsequently queue time. 156 The major question is the cost effectiveness involved in any of the alternatives. The viable approach is to identify what combination of the above actions will reduce total delay to a tolerable level at the least cost. Hence a linear optimization program will be employed in chapter VI. The sensitivity of queue waiting time to a reduction in berthing time is illustrated by the family of curves shown in figure (6-3). When berth service time is reduced by 20%, que waiting time drops by 38%. A reduction of berth service time by 40% yields 78% reduction in queue waiting time. Hence the family of curves in figure (6-3) provide the basis for setting alternative optimization constraints. REFERENCES Wohl and Martin: Traffic Systems Analysis for Engineers and Planners. McGraw-Hill, New York. Ericksen, Stian: Simulation of Receiving, Storing and Loading General Cargo. University of Michigan, Department of Naval Architecture and Marine Engineering, Ann Arbor, Michigan 48104. Rossa, 6.: Investigation of the Distribution of Ship Arrivals in a Line Service. Seewirthschaft; Berlin, East Germany. Frankel, E. G., P. Wilmes and K. Chelst: Simulation of Multipurpose Port and Multipurpose Offshore Facilitflgg. Publication Off- shore Technology Conference. Dallas, Texas. Erickson, S.: Optimum Capacity of Ships and Port Terminals. University of Michigan, Department of Naval Architecture and Marine Engineering, Ann Arbor, Michigan 48104. Collier, P. I.: A Simulation Model for Ports Management Training. Dock and Harbour Authority, Foxlow Publications 19 Harcourt Street; London NIH 2AX; England. 157 CHAPTER VI INVESTMENT OPTIMIZATION The overall approach is to determine the optimum combination of port resources which will accommodate the specified thoroughput at the minimum annual cost. The following investment options will be considered: - construction of additional units of berths - construction of new warehouses - acquisition of additional units of gantry cranes, fork lifts, yard transfer equipment, and floating cranes - purchase of additional number of tugs for pilotage and towage services ° purchase of additional train cars and power units for improving cargo delivery ° acquisition of additional land for both anchorage facilities and open storage area - investment in signal system and traffic control devices - increase in the number of supervisory and clerical staff assigned to warehouses. The emphasis is on optimization of the entire port system rather than any of the subsystems mentioned above. A linear programming model can be employed to determine the optimum combination of the above alternatives which will maximize total annual thoroughput at minimum total annual cost. Hence time and tonnage constraints are very im- portant factors. 158 159 6.1 The Objective Function For this program, the total annual cost of port operation and ownership was written as the objective function: Minimize Z; where z = axln + bX3N + dx4T; + ex5(T§) + (Xe/T6 + CV6) + (x7/T7 + CV7) + x8/T8 + Cv8) + (Xe/T9 + CV9) + (XIO/Tlo + CV10) + (X11’T11 + CV11) + (Xlz/le T CV12) + (X13/T13*'CV13) + (x14/T14 + CV14) + (x15/T15 + CV15) T (X16/T16 + CV16) + (x17/T17 T CV17). + (X18/T18 + CV18) + cvx19 + CVX20 where Xi = capital cost of ith investment Ti = economic service life of the ith investment Cvi = variable cost associated with ith investment or labor T; = % of total cargo tonnage which moves through transit warehouse T: = % of total cargo tonnage which moves through inner harbor facilities a, b, d and e are cost coefficients defined in table (6.2). N = total number of ships entering Lagos port in one year. N is not a variable for a specific year Tt = thoroughput of cargo in one year (tonnes). Tt is not a variable for a specific year The above equation can be simplified as shown below: 160 Minimize z = ax N + bX N + dX (T') + ex (Ta) 4 t 5 t 1 3 T CT6X6 T CT7X7 T CT8x8 T CT9X9 X 10 C ,X + C X T11 11. T12 X 15 X 19 + C T 10 T 12 T CT 3x13 1 X 16 X 20 T C C 15'T CT 16 T CT 7X17 X + T14 14 T 1 X 18 + C + C + C T 18 T 19 T 20 where CT = (Xi/Ti + Cvi) = Total annual cost per unit of 1 investment or labor 6.2 Determination of Optimization Constraints The data in table (6.1) was obtained from the work study department of the Lagos port. It provided the basis for estimating additional units of investment as a function of ship service delay. A number of equations were developed to express the reduction in delay as a function of increments in equipment and storage facilities. Using the assumption .that these relationships are linear, the following equations are gener- ated: From Column 1: X9 + X19 + 2X16 + X13 + X10 + X11 = 12 days --Equation (1) From Column 2: 2X9 + 2X19 + 4X16 + X13 + X10 + X11 = 10.6 days --Equation (2) From Column 3: 2X9 + 2X19 + 8X16 + X13 + X10 + X11 = 9.2 days --Equation (3) Subtracting Equation 2 from Equation 3 4X16 = 1.4 days X16 = .7 days 161 From Column 4: 9 + 2X19 + 8X16 13 + X10 + X11 = 7.8 days --Equation (4) Subtracting Equation 3 from Equation 4 X13 = 1.4 days 2X +2X Detailed discussion of the optimization constraint for each state vari- able are given in table (6-2). From Column 5: 2X9 + 2X19 + 8X16 + 2X13 +2X10 + X11 = 7.5 days --Equation (5) Subtract Equation 4 from Equation 5 X10 = O 3 From Column 6: 2X9 + 2X19 + 8X16 +2X13 +2X10 + 2X11 = 7.0 days --Equation (6) Subtract Equation 5 from Equation 6 X11 = .5 days From Column 7: + 2x' 3x7 + 4x' + 8X. + 2x' + 3x 11 9 19 16 13 10 Subtract Equation 6 from Equation 7 = 2.2 days --Equation (8) 4.8 days --Equation (7) x9 T 2x19 T X1o From Column 8: ' + 6x' + 8X. +-2x' ' ' = 3.6 days —-Equation (9) 9 19 16 13 T 3x1o T 2x11 Subtract Eauation 9 from Equation 7 3X 2X19 = 1.2 days '3. X19 = .6 days Substituting values of X10 and 519 in Equation 8 (2.2 - 1.2 - .3) X9 0.7 days Hence the ship unloading time (X21) can be expressed as follows: 162 X I 21 - f(Gantry cranes, labor gang, forklift, warehouse, open space storage area, train units) - K - .07X9/X8 - .3X10/X8 - .5X11/X8 1.4X13/X8 - .2X16/X8 - .9X19/X8 ...: (D x I 21 Since these values are based on the assumption that the relation- ships between each variable are the ship unloading time on independent, and that all relationships are linear, the use of these results should be limited to values near those in table (6.1). mmmccoh xmmm :o comma xvzumw 163 mcou oon u avzm can mmmccoh muogm>< .mommH mcou ooom u awnm cog mmmcco» Ham; ”>ax .>u_go3H=< ago; :NHmeHz ”muczom mzcu when mANc mxmn mxac maau mxmu mxmc HmcwumoH new mcwumoHcav m.m m.e oo.~ m.N m.~ N.m m.oH NH wow>gmm amgm mmmgm>< N N N H H H H H chaahv ocmsawaau cowumusoqmcmgh m m N N H H H H Hmmgum ooHv swam mmogoum mumam cmao N N N N N H H H mmzogmgmz N m m m m m e N HNHH Hook 0 e N N N N N H 9:8 .893 m m N N N N N H mmcmgo Ngucmw goamm am mooczommm mo cowamcmnEou can Ham5apacm we gmnsaz mmucsommm toAh mwzmum aHzm mw< mo >HH>HHHmzmm .AHuov mpnmh 164 Table (6-2). OPTIMIZATION CONSTRAINTS VAETREEES Definitions Constraints and Remarks N Number of ships N = f(Throughput) cleared in the study = Tt/tav. port in a year. where tav = average load carried per ship T1: = total tonnage through the port in one year. a Average cost of one The 1978 demurage average was ship/unit waiting $105.00 per hour. time in queue. X1 Average ship waiting Determined from traffic simulation, ' - - i.e. time in the Delay = 30 hrs. if X8 = 64 queue _ . _ Delay - 2.85 hrs. if X8 - 70 2 X1 3_30 - 4.4(X8 - 64) - 1.5X7 where 1.5x7 is the tug transit time. X2 Total time Spent by X2 = XI + X3 a 5m" T" the.”“""" '. x > 30 - 4.4(x - 64)- 1.5x +x 2 —- 8 7 3 X3 Service time in berth. X3 = X2 - X1 b Average cost of $200.00 per hour 1978 ship berthing time. x4 Average cargo waiting Treat as a fixed cost (i.e. con- time in transit ware- stant cost). Shippers are allowed house. 4 days of grace for storage. c Average cost of one $2.25 per ton/hr. 1978 cargo unit waiting time/warehouse. 165 Table (6.2). Continued STATE VARIABLES Definitions Constraints and Remarks x6 Land allocation invest- ment for anchorage facilities X f(Number of berths) 2 X6 can be replaced in the optimization equation by 2X8 6 Number of tugs The constraint specifies a daily Tug/ship ratio of .05, i.e. _ .05N _ x3 " "21T1-5 (‘36? x7) where X3 service time at berth X21 = unloading time 1.5 transit tug time from queue to berth and back to the queue. Number of berths Determined from the simulation at a specified service rate. However X/X8u = p < 1 ship arrival rate per hour for the entire port ship service rate per unit berth where A t: ll X Number of Gantry cranes X9 = f(Tonnage through the port in a day) Lt. 1 356 'TFF The capacity of Gantry crane under the 8 hour work day = 24 tons. 356 day excludes all public holidays in Nigeria. '. X9 = 166 Table (6.2). Continued SASTABLES Definitions Constraints and Remarks X10 Open space storage area X10 = f(% of indirect cargo ton- in acres. nage through open storage areas per day, rate at which cargo moves out per day in open storage area). 0 10 - 356 356 20 I ('I' = .04 T 1 333- t 7)- acres where, 10% of indirect cargo moves through open storage. 60% of the above tonnage moves out per day. 20 tons is stored in one acre. 11 Land Transportation X11 - f (Tt) Investment Train - . .( ‘ -33 It 1/150 K units requ1red). 556' where 83 = % of the direct deliv- ery cargo moved out of the port by trucks and rail. K = % of the above moved by rail (i.e. 20%) 1500 tons is the capacity of train unit operated in the port 167 Table (6.2). Continued EAQTABLES Definitions Constraints and Remarks X12 Number of Storage Warehouses. x12 = f (Tt) = .023 Tt _ (.60)(.023) Tt 366 356 .001 m2 2 ’011 Tt / .001 meter 356 where .023 is the proportion of Tt which moves through storage warehouse. 60% of the above cargo is cleared per day. .001tons/meter2 = warehousing storage standard X13 Number of Transit X13 = f (Tt) Warehouses = f Tt 356 =[:°7 1L - (.60)§.07) Tt] 356 3 6 m2 .001 where .07 is the proportion of indirect cargo moving through transit warehouse. ..001t0ns/m2 is the warehousing storage X14 Dredging Investment standard. X14 = f (length of channel, number of berths) X14 = f (l + 250X8) X14 = (4633 + 250X8) where l = length of channel = 2.5 naut- ical miles (4633 meters). 250 = length of berth. Table (6.2). Continued 168 STATE . . . VARIABLES Definitions Constraints and Remarks X15 Signals Investment where $30 is the cost of ship signals per entry. N = total number of ships in a year. X16 Number of Fork Lifts X16 = f (Tt) = 1 Tt THE’ 3??? where 24 tons per 8 hour day is the maximum capacity handled by a fork lift. (Union restrictions)- X17 Number of Yard Transfer Equipment X17 = f (Tt) T113. 40’ 356 where 40 tons is the maximum capacity handled by equipment per 8 hour day X18 Number of Floating Cranes X=f_u_ 356 =-2_1_ 356 where .2 is an acceptable floating crane to ship ratio necessary to service a sudden increase in demand. Treat as fixed cost. 18 X19 Number of Labor Gangs X19 - f (Tt) 1 it. :nr’ 3 6 where 24 Tons = daily union productiv- ity limit per gang. 169 Table (6.2). Continued 3A3T§BLES Definitions Constraints and Remarks X20 Number of Supervis- X20 = f (Tt) ory Staff = 1 .12 ITS' 356 = 7 Tt E? where maximum daily tonnage processed by on supervisory staff = 1.5 tons per day. X21 Ship Unloading Time X f (X X X X 13’ 16’ x19) .3x10/x8 .sxn/x8 - 1.4x13/x8 .2x16/x8 - .9x19/x8 where K = 15 days (1.1, . . . . .5 are delay minimiza- tion achieved by introduction of one additional unit of equipment or labor in a berth.) The coefficients are derived from work- study data. 9’ X10: 11’ K/2 - 0.7x9/x8 - 21 x21 Safety Constraints 01X9/X8_<_2 0 i X19/X8 _<_ 3 .berth. O < X16/X8 < 2X19 Maximum of two Gantry Cranes per one berth. Maximum of three labor gangs per one (20 men in each gang) 170 Table (6.3). SUMMARY OF OPTIMIZATION CONSTRAINTS VARIABLES CONSTRAINTS N N = Tt tav X1 x1 3_30 - 4.4(X8 - 64) - 1.5x7 X2 X2 = X1 + X3 X2 3_30 - 4.4(X8 - 64) - 1.5x7 + X3 X3 x3 T X2 ' x1 X4 Treat as constant cost. X6 X6 = 2X8 Replace X6 by2 X8 in optimization equation X X = X + 1.5 .05N 7 3 21 Tfifi§" X7 X8 Determined from simulation T X X = .04 t 9 9 -§§— X10 X10 = .002 Tt 3 6 X11 X11 = .166 Tt 356 1500 x x = 03 T meter2 12 12 ' t 171 Table (6.3). Continued VARIABLES CONSTRAINTS X X = .08 T meter2 13 12 t X14 X14 = (4633 + 250X8) X15 X15 = 30N X16 X16 = .04 Tt/356 X17 X17 = .03 .IE. 356 X18 X18 T '2 32% 356 356 ' .2X16/X8 ' .9X19/X8 where K = 15 days. 172 where K/2 is the maximum berthing time (15 days for loading and unloading). Table (6-2) discusses the criteria for establishment of other con- straints. The simulation output provided inputs which created berthing and service time constraints. 6.3 The Optimization Process When the various cost coefficients are introduced the objective function changes to the form: Minimize Z Z = 105x N + 200x 1 2 + 5,128 X6 + 179,56OX N + 2.25X4 ('T7Tt) 7 + 169.173X8 + 21,9OOX9 + 46,6OOX10 + 39,829X11 + 45X12 + 76X + 240x 13 14 T x15 + 2,773X + 15,025X17 + 180,000X . 16 + 54,000x 18 19 + 3600X20 the number of ships entering the port in one year is where N a constant for the specific year. .4 ll total cargo tonnage through the port in one year is also a constant for the specific year. The next step is to rewrite the variables in terms of X8 and Tt as dis- cussed in table (6.2). The objective function changes to the form T below: 173 Minimize Z Z = 105NX1 + ZOONX3 + 2.25X4(. 179,56OX7 + 169,173X + 21,900(.04)Tt 356' 17Tt) + 5,128X6 + 8 + 46,600(.002)Tt + 39,829(.166)Tt '356’ 1500 x 356 45(.O3Tt ) + 76(.08Tt) + 240(4633 + 250x 30N + 2773(. O4)T + 15,025(.O3)Tt + 8) + _JL. 356 356 + 180,000(.2) Tt + 54 ,000(. 04 )T t (3000) (3567 356' + 3600(.66)Tt 356 where-Tt = the total tonnage in a specific year (Tt for 1978 = 1.12 x 106 tonnes) N = 5000 ships for 1978. Simplifying the objective function reduces to: Minimize Z Z = 105X1N + 200X3N + .383X4Tt + 5,128X6 + 179 ,560X7 + 169,173X8 + 2.46Tt + .26Tt + .01Tt + 1.35Tt + 6.1Tt + 240(4633 + 250X8) + 30N + .3Tt + 1.27Tt + .03Tt + 6.07Tt + 6.7Tt The next step in the reduction process is to write the objective func- tion in terms of X8 and Tt’ i.e. Minimize Z 105x N + ZOOX N + .383X T + 10,256X 1 3 4 t 8 169,173X8 + (2.46 + .26 + .01 + 1.35 + 6.1 + .3 + 179,560X7 + + 1. 27 + .03 + 6.07 + 6. 7) Tt + 30N + 1,111,920 + 60 ,ooox8 174 Simplifying the objective function reduces to Minimize N + 200x N + .383X Z = IOSX + 179,560X7 + 239,426X8 1 3 4 + 24.55Tt + 30N + 1,111,920. In the above objective function X4 (average cargo waiting time in trans- it warehouse) is considered a fixed cost because shippers are allowed 3 days of grace for transit storage. This means that according to the existing contract the minimum value of X4 is 72 hours. Hence X4 is set at this value and the associated cost expressed in terms of Tt' When the value of N (N = 5000) is substituted in X1 and X3 the objective function reduces to: Minimize Z = 525,000X1 + 1,000,000X3 + 179,560X7 + 239,429X8 + 52Tt + 30N + E where E is the investment level in berthing equipment and labor. Investment constraints were generated from the work study data in table (6-1). These constraints relate X3 (berthing time) to the in- crease in number of equipment and labor gangs. The data obtained from table (6-4) were plotted as shown in figure (6-2). A linear relation- ship was observed between investment cost of berthing equipment and labor, and berthing time X3. Hence X3 can be expressed as a function of the level of investment, i.e. X - E) 3 2-’T2 " A3 (Emax where Emax is the maximum annual investment in berthing equipment and labor. A3 is the slope of the curve in figure (6-2). A2 is the maximum berthing time associated with an annual invest- ment level E. 175 unaccoh xmma no woman Huaum « mcou oon u q_sm Ema mmmccou mmmcm>< Hmcou ooom u anm can unaccoMHNma H>mx :opumcHgEou somm :HH3 NHN.HHH.H Nem.moo.H NHN.HNN NHo.NHN NH¢.oNN HH¢.oom HNN.HN¢ NoH.Ho¢ H coHaHuoHH< Haas upmm>cH HN:::< Hmuop mxmc mama mxmc what mxmu what what mxmu HmcHuNoH + mcwcmochv H.N N.H oo.N H.N H.N N.H H.oH NH auH>amH NHHH aaaaa>< Hanghv NHH.HN NHH.HN NHH.HN HNN.HN HNN.HN HNH.HH HNN.HH HNN.HN H Hmoa HeasaHzaN N N N H H H H H .6: 56HH coHHaHaoamgmaH Hmmgum ooHv oom.HNH oom.HmH LQON.NH OON.HH ooo.ca OOH.HH ooo.oa .ooo.o¢ H Hmoa Nata amaaoom m m N N H H H H .o: EmHH mumam :mgo ooo.cN¢ ooo.ome ooo.ow¢ ooo.oma ooo.om¢ ooo.oaN ooo.caN ooo.oeN H Hmou N N N N N H H H .o: ENHH amsogaamz NNH.NN HNH.NN «NH.NN «HH.NN HNH.NN «NH.NN NHo.HH HNN.N H Hmou N N N N N N a N .6: 26HH HHHH Naoa ooo.eNN coo.HHN ooo.mcH ooo.moH coo.NoH ooo.moH ooc.NoH ooo.em H Hmou H H N N N N N H .oz saHH Hana toaaH ooN.HH ooN.HH oom.m¢ ooN.Ha oom.mv oom.mq oom.ma oom.HN H Hmou H H N N N N N H .6: 56HH Hagaau Hchmo HzH HHH HHH HNH Hay HNH HNH HHH<2NNHH< guaam Ha moogaomom Ho :oHHmcHnEou ccm acmsaHacm Ho gmaszz mmugzommm “can Hmmm aHzm uu< no >hH>HhHmzmm .HH-HV aHHmH 176 . 33.225 a m: z— .58 59.5ng @5253 Illlll a ...». .H. 2.. ... ... ...: §.H.B¢HH.N8°.SN.H : . . n86. .- an n fix? .9... RN u a 3.39 ”53.. 92 238.3.ch :— mm:>:Hmzwm 473 0.53... 88.8... 88.8, .2 H: (sxvo NI.) awn 91411112139 *—— 177 A knowledge of ship turnaround time is also necessary to describe total ship delay. This time component is based on port investment policy and maximum turnaround time tolerable to the shippers. The turnaround time can be written as follows: where X1 = waiting time in queue X3 = total berthing time (unloading + loading) X7 = number of tugs and associated transit time to and from head of the queue to berth. .< ll maximum turnaround time tolerable to the shippers. Linear Programming,Model: The package employed is a version of the North Western University Vogelback system. In the Michigan State Computer Center it is identified as APLIB, LTT 5640, P*LP. This package requires a consistent naming of independent variables. Hence. the variables with alpha codes were renamed: T=X9 NTXlo Is=x11 When these identities are substituted, the objective function becomes: Z = 525,000X1 + 1,000,000X + 239.429X8 + 52X9 + 30X10 + X + 179,56OX 3 7 11 Care should be taken not to confuse these variables with the initial meaning of X9, X10, X11 in section 6.1. As shown in the computer printout three alternative port invest¢ ment combinations were optimized. These alternatives will be identified as cases 1, 2, and 3. Each of these alternatives were run for the 1978 178 ship and cargo traffic demand. In case 1 the existing berth service time of 199 hours per ship is kept constant. Fifteen new berths are created in addition to the existing 39 berths. This brings the total number of berths to 54. The solution to the linear program results in a queue waiting time for this alternative of 6.4 hours per ship. The second alternative proposes a 25% reduction in existing berth service time and the creation of only ten new berths. This reduction in berth service time can be achieved by increasing the investment level in equipment (i.e. the X11 variable in the optimization con- straints). Table 6.4 illustrates the impact of various investment levels on the ship service time. In case 2, the queue waiting time is 4.4 hours which is 30% lower than case 1. In addition, the annual cost to the port (2) in case 1 is 23% higher than case 2. Hence alternative 1 is not economically viable to service the short term traffic demand of the port. The third alternative results in a lower cost than case 2. In this alternative, only five additional berths are created and the berth service rate is reduced to 120 hours (40% reduction). The queue wait- ing time for this solution is reduced to 1.24 hours. This is due to the associated cost savings resulting from the reduced waiting time. The annual cost '2' is 20% lower than the value obtained in case 2, and is the recommended policy for dealing with short term traffic and logistical problems of the port. In section 6.3b specific investments are described to determine the optimum combination of port subsystems to meet 1990 ship and cargo traffic demand. 179 £953 .3 5932 ‘II .5332... S: .9... 87.... ...dl ll... asHH ouHseoH enema H5882. NHNH .2: 8H . «I: motion 5..: ou.seoH eueoa HHHHHHNN >5. .3253 N3: Hz: 8:53. :2: 3 5:88.. 2 Hz: 3:2... N8 3 EEHHzNH .38 2.5: 5% 5? fl ~ :3 N '(‘Ivmuv 111001 "$811) 31411 sumvn am 0 m N 180 Table (6-5). SUMMARY OF OPTIMIZATION RESULTS (1978 DEMAND) CASE 1 VARIABLES DEFINITION OPTIMUM INVESTMENT REQUIREMENTS X Waiting time in 1 the Queue 4.4 hours X Berthing time 3 (unloading + loading) 48 hours X Cargo Waiting time 4 in transit warehouse 72 hours X6 Land allocation investment for 128 units anchorage facilities X7 Number of Tugs 21 units X8 Number of Berths 64 X Number of Gantry 9 Cranes 71 X Open space storage 10 area in acres 7 acres X11 Train unit Invest- 2 unit ment trains X12 Number of Storage 6 Warehouses X13 Transit Warehouses 14 X14 Dredging investment $4,533,250.00 X15 Signals investment $150,000 X Number of fork lifts 132 units 181 Table (6-5). Continued VARIABLES DEFINITION OPTIMUM INVESTMENT REQUIREMENTS X Number of Yard . 17 Transfer Equipment 80 units X Number of Floating . 18 Cranes 3 units X Number of Labor 19 Gangs 132 gangs X20 Number of Supervisory 2097 and Clerical Staff 182 Table (6-6). SUMMARY OF OPTIMIZATION RESULTS (1978 DEMAND) CASE 2. VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 1 REQUIREMENTS X1 Waiting time in 4.4 hours -2.0 hours the Queue X3 Berthing time 150 hours -49 hours (unloading + loading) X4 Cargo waiting time 72 hours -- in transit warehouse X6 Land allocation 98 units -10 units investment for anchorage facilities X7 Number of Tugs 18 units —11 units X8 Number of Berths 49 -5 berths X9 Number of Gantry 90 units +18 Cranes ! X10 Open space storage 1 9 acres +2 area in acres X11 Train unit Invest- 2 unit trains -- ment X12 Number of Storage 8 +2 Warehouses X13 ~ Transit Warehouses 18 +4 X14 Dredging investment $4,220,750.00 -$312,500 X15 Signals investment $150,000.00 -- X Number of fork lifts 160 units +20 183 Table (6-6). Continued VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 1 REQUIREMENTS X17 Number of Yard 100 units +20 Transfer Equipment X18 Number of Floating 3 units -- Cranes X19 Number of Labor 160 +28 gangs Gangs = 560 men X20 Number of Supervisory 2600 . +503 men and Clerical Staff 184 Table (6-7). SUMMARY OF OPTIMIZATION RESULTS (1978 DEMAND) CASE 3 VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 1 REQUIREMENTS X1 Waiting time in 1.24 -5.16 the Queue X3 Berthing time 120 -79 (unloading + loading) X4 Cargo Waiting time 72 -- in transit warehouse X6 Land allocation 88 units -20 investment for anchorage facilities x7 Number of Tugs 13 -16 X8 Number of Berths 44 -10 X9 Number of Gantry 100 units +28 Cranes X10 Open space storage 12 acres +5 area in acres X11 Train unit Invest- 3 unit trains +1 ment X12 Number of Storage 10 +4 Warehouses X13 Transit Warehouses 20 +6 X14 Dredging investment $3,908,250.00 -615,000.00 X15 Signals investment $150,000.00 -- X Number of fork lifts 184 +52 185 Table (6-7). Continued VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 1 REQUIREMENTS X17 Number of Yard 112 +32 Transfer Equipment X18 Number of Floating 3 units -- Cranes x19 Number of Labor 180 +48 gangs = 960 men Gangs X20 Number of Supervisory 2936 +839 men and Clerical Staff 186 Table (6-8). COST EFFECTIVENESS OF 1978 (SHORT TERM) ALTERNATIVES ALTERNATIVES MINIMUM VALUE SAVINGS % SAVINGS OF Z (S) ($) Case 1 280,498,739.00 base -- Case 2 217,044,044.00 63,454,695.00 23% Case 3* 194,025,418.00 83,473,321.00 31% NB: Alternative No. 3 is recommended on the basis of minimum cost and reduction in total delay to ships and cargo. 187 Port Investment Projections: The optimum combination of port invest- ments required to service future ship and cargo demand can be determined by projection. In this projection there are five major steps: - Forecasting of ship traffic volume for the future year based in historical data. This forecast assumes that there is no radical change in economic growth. ' Forecast of the cargo tonnage for the future year. ' Estimate of ship arrival rate (IPRATE) for the future year. - Simulation of the future year ship traffic. - Optimization of port investment requirements for the future. Ship traffic volume expected to enter the port of Lagos in 1990 can be predicted from figure (2-8) by applying the following equation: nggo = ~1978 + a(v, - v1) ------------------------------- (i) where "1990 = predicted number of ships in 1990 N1978 = number of ships which entered the port in 1978 a = the slope of the demand curve Y2 = projected year Y1 = base year. Figure (2-8) illustrates that there has been a linear growth rate in ship volume between 1974-1978. This growth rate is expected to con- tinue because of the stability in government. Hence the slope of the demand line can be computed as follows: a = N - N .1228 1912 1978 - 1975 = 5,800 - 5,200 0! II 299 188 When this value is substituted in equation (i) above the volume of ships expected in 1990 can be determined, i.e. N = 5,800 + 200 (1990 - 1978) 1990 5,800 + 2,400 8,200 ships. The next step is to forecast the cargo thoroughput in 1990. This estimate is based on the fact that cargo thoroughput is directly pro- portional to the number of ships calling at the port. i.e. When N = 5,800 ships, total tonnage (Tt) = 1.12 x 106 tonnes . 6 8,200 ships, T = 1.12 x 10 x 8 200 tonnes t 5.7300 1 when N . T _ 6 .. t1990 - 1.58 x 10 tonnes. Another important variable that should be determined is the ex- pected ship arrival rate for 1990. This estimate is based on a rational assumption that ship arrival rate is directly proportional to the number of ships entering the port. i.e. When N = 5,800, arrival rate (IPRATE) = .35 ship/hr. '. when N = 8,200, IPRATE = .35 x 8,200 Ship/hr. 5,800 arrival rate for 1990 = .49 ship/hr. This 1990 ship arrival rate was introduced in the simulation model to generate expected ship waiting time and queue length, with the 1990 ship traffic volume and tonnage specified as optimization constraints. 189 The family of curves shown in figure (6-3) were developed from the results of simulation programs for the projected 1990 ship arrival rates. Ship berthing time was varied from 199 hours to 150 hours and 120 hours (i.e. a reduction of 25% and 40% respectively). Optimiza- tion constraints relating queue waiting time X1 and the number of berths X8 was derived separately for each curve. These alternative optimization programs are identified as cases 4, 5, and 6. As shown in the problem statements, the cost of the various items in the objec- tive function for 1978 was increased by 10% to reflect the expected increase in unit costs by 1990. In case 4 the existing berth service time is maintained at the existing rate of 199 hours per ship. The total number of berths is increased to 64. The queue waiting time for this case is 3 hours. In case 5 the berth service time is reduced to 150 hours per ship and total number of berths is reducted to 59, yielding a queue waiting time of 4.8 hours. In case 6, the berth service time is further re- duced to 120 hours and only 10 new berths are created. The queue waiting time for this case is 5 hours, which is 20 minutes higher than case 5. However, the annual cost savings of 20% over case 5 offsets this increase in waiting time. Hence this last alternative is recom- mended as a program for meeting the increased ship traffic and cargo tonnage in 1990. 19() I. a4 a 283 [I ’ ’ c.m u ago—m Ace-ausoa no: .2: cNH .. I I I 8:. auwtflm 550m .5333: $3 J.:. 2.; a as.» ouH>eom sauce .9... 8H .. 8: oeH>eeH eaten HeHHHqu 35:38:; 625.3 :33 us: mug—gum 2.5%. .3 52.25%. A: at: $22,229.35 “8 >:>_:mzum >9. £933 .3 «mg: I .H Hm m. IA: / canoe; r//// a. 0' .313 95m: an 7H ..H In F. a; ('saH) 1vnxuuv «1001 303 3w11 9N111vn 300 .3 6mm Table (6-9). 191 SUMMARY OF OPTIMIZATION RESULTS (1990 DEMAND) CASE 4 VARIABLES DEFINITION OPTIMUM INVESTMENT REQUIREMENTS X1 Waiting time in 3 hours the Queue x3 Berthing time 199 hours (unloading + loading) X4 Cargo Waiting time 72 hours in transit warehouse X6 Land aIIocation 128 units investment for anchorage faciTities X7 Number of Tugs 20 units X8 Number of Berths 64, X9 Number of Gantry 184 Cranes X10 Open space storage 14 acres area in acres X11 Train unit Invest- 4 unit trains ment X12 Number of Storage 12 Warehouses X13 Transit Warehouses 22 X14 Dredging investment $5,158,250.00 X15 SignaIs investment $270,600.00 X Number of fork Tifts 190 units 192 Table (6-9). Continued VARIABLES DEFINITION OPTIMUM INVESTMENT REQUIREMENTS x17 Number of Yard 120 units Transfer Equipment X18 Number of Floating 5 units Cranes X19 Number of Labor Gangs 190 X20 Number of Supervisory 3,108 and Clerical Staff Table (6-10). 193 SUMMARY OF OPTIMIZATION RESULTS (1990 DEMAND) CASE 5 VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 4 REQUIREMENTS X1 Waiting time in 4.8 hours +1.8 the Queue X3 Berthing time 150 hours -49 (unloading + loading) X4 Cargo Waiting time 72 hours -- in transit warehouse X6 Land allocation 118 units +10 investment for anchorage facilities X7 Number of Tugs 24 units +4 X8 Number of Berths 59 -5 X9 Number of Gantry 230 units +46 Cranes X10 Open space storage 18 acres +4 area in acres X11 Train unit Invest- 4 units -- ment X12 Number of Storage 15 units +3 Warehouses X13 Transit Warehouses 28 +6 X14 Dredging investment $4,845,750.00 -$312,500.00 X15 Signals investment $270,600.00 -- X Number of fork lifts 238 units +48 194 Table (6-10). Continued VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 4 REQUIREMENTS X17 Number of Yard 150 +30 Transfer Equipment X18 Number of Floating 5 units -- Cranes X19 Number of Labor 240 +50 gangs Gangs i.e. 1000 men X20 Number of Supervisory 3800 +692 men and Clerical Staff Table (6-11). 195 SUMMARY OF OPTIMIZATION RESULTS (1990 DEMAND) CASE 6 VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 4 REQUIREMENTS X1 Waiting time in 5 hours +2 the Queue X3 Berthing time 120 hours -79 (unloading + loading) X4 Cargo Waiting time 72 hours -- in transit warehouse x6 Land allocation 98 units -30 investment for anchorage facilities X7 Number of Tugs 20 units -- X8 Number of Berths 49 -15 X9 Number of Gantry 250 +66 Cranes x10 Open space storage 22 acres +8 area in acres X11 Train unit Invest- 4 unit trains -- ment X12 Number of Storage 18 units +6 Warehouses X13 Transit Warehouses 34 +12 X14 Dredging investment $4,220,750.00 -$937.500.00 X15 Signals investment $270,600.00 -- X Number of fork lifts 286 units +96 196 Table (6-11). Continued VARIABLES DEFINITION OPTIMUM CHANGE FROM INVESTMENT CASE 4 REQUIREMENTS X17 Number of Yard 180 +60 Transfer Equipment X18 Number of Floating 5 units -- Cranes X19 Number of Labor 280 +90 gangs Gangs i.e. 1800 men X20 Number of Supervisory 4560 +1,452 men and Clerical Staff Table (6-12). Ia COST 197 EFFECTIVENESS OF 1990 (LONG TERM) ALTERNATIVES ALTERNATIVE MINIMUM VALUE SAVINGS IN % SAVINGS OF 2 ($) (3) Case 4 $321,748,474.00 base -- Case 5 $267,930,186.00 $53,818,288 17% Case 6* $232,411,966.003 $89,336,508 28% I Case 6 is recommended for 1990 demand based on minimum annual cost and.performance. 198 6.4 Traffic and Logistical Operations Survey In addition to the quantitative analysis, a survey was designed and condusted to identify those factors which are considered to be inadequate for present port operations and to obtain information on logistical systems. The survey responses will help the planner to interpret the results of the linear programming models. The methodology adopted in this survey considered three major groups: a Port traffic officers - Port operations officers - Shippers and freight forwarders. In order to generate more reliable information, only officers in responsible charge were interviewed. The officers were directly involved in ship and cargo processing. This was done to ensure that experimental subjects have adequate experience in the day- to-day problems in their various areas. Personal calls were made to each of the selected officers to explain the basic coding format of the survey forms. This is the reason for the high state of response achieved. Table (6.13) on the next page summarizes the responses received from various survey categories. Traffic officers responded 5% more than operational officers or shippers. TABLE (6.13) RESPONSE TO LOGISTICAL OPERATIONS SURVEY FORMS 199 CATEGORY DESCRIPTION NUMBER OF FORMS RESPONSE % RESPONSE A Traffic 20 16 80 Officers B Operations 20 15 75 Officers C Shippers 20 15 75 As Shown in tables (6.14-1), (6.14-2) and (6.14-3), analysis of the survey forms administered to each category are summarized. Traffic officers and operations officers agree that the following facilities are inadequate to service the existing demand. - Berths - Gantry cranes and fork lifts - Yard transfer equipment - Warehouses - Signals, tugs and pilotage. Traffic officers and operations officers remarked that shippers' un- willingness to clear cargo on time is one of the principal causes of cargo delay within the port system. On the other hand, shippers argue that poor cargo tracing and lack of information about berthing schedules are principal causes of cargo delay. In addition, a majority of shippers see the custom 200 clearing process in Lagos port as complicated and outmoded. Tables (6.14-1) through (6.14-3) illustrate the rating of each subsystem by the different groups. The highlights of the ranking are shown below: Ist Critical Average 2nd Critical Average Group Subsystem Score Subsystem Score (a) Traffic officers Gantry cranes 3.4 Yard Transfer 3.7 8 forklifts equipment (b) Port operations Berths 2.9 Cargo clearing 3.3 officers process (c) Shippers Berths 1.6 Gantry cranes 2.9 A close examination of the above results indicate that there is a con- census between the two groups (operations officers and shippers) that the Number of berths at the port is not adequate for the present demand. Traffic officers and Shippers also agree that the number of gantry cranes and forklifts assigned to each berth are not adequate. All the three groups rated signals, pilotage and towage systems as satis- factory. However group bias was recognized in the question of cargo clear- ing. Port officials tended to attribute the cause of cargo delay to shippers unwillingness to clear goods in time. On the other hand, shippers blamed port management for the complicated custom requirements involved in cargo clearing. In the writer's opinion both parties share equal blame for the time lag involved in cargo clearing. The solution lies in effective communication between the two parties. 201 Table (6.14-1). ANALYSIS OF TRAFFIC AND LOGISTICAL OPERATIONS SURVEY FORMS. Category A: Traffic Officers. 1. Rank the operational adequacy of the following port subsystems. Assessing values (See KEY) to any of the subsystems order. a) Signal System b) Pilotage and Tugs c) Berths d) Gantry Cranes and Forklifts e) Labor Gangs per berth f) Warehouses (Transit sheds) g) Cargo clearing process h) Yard transfer equipment KS : Excellent I 8 Good ' 6 1 2 [321 [:Jul [:I--2 [322 El" m2« Qua Elna Average Very Good - 7 Above Average - 5 Below Average 2. Rank the major causes of ship delay in the port. a) Poor signal, tugs and pilotage b) Lack of sufficient number of berths c) Lack of sufficient number of Gantry Cranes and Forklifts (loading and unloading)__ d) Insufficient number of receiving ware- houses (Transit sheds) e) Labor problems I: [:1 [:1 [:1 D 5 N H 3 4 5 6 7 8 -- 2 -- 2 2 3 -- 2 -- 3 5 2 3 2 I 3 3 -- 4 3 -- 3 -- -- 2 5 3 3 1 -- 5 2 5 2 -- -- 5 3 2 1 ~- 2 5 2 1 -- 1 . 4 Poor - 2 - 3 Bad 8 1 12345 5------2 s----44 113--9 N3 Rank should run from 1 to 5 (i.e. major cause a 5, minor cause . 1) KEY: Major cause - 5, Minor cause - 1, Not as a Factor - O 3. Rank the factors which cause cargo delay in the port of Lagos. a) Poor handling equipment (Cranes, forklifts conveyors) 0) Lack of contain transfer equipment c) Poor record keeping in transit sheds ____ d) Complicated custom clearing process e) Shipper's unwillingness to clear goods on time t: D a a [:1 O 1 2 3 4 s 21----7 5 4113 71213 3 --1 18 ----338 N8 Ranks should run from 1 to 5 (major factor - 5, minor factor . 1) Major factor *1 5, Minor factor = 1, No factor - O 4. Signals, tugs and pilotage services in the port can best be described as: a) Excel lent b) Good c) Average d) Poor DUDE 2 9 4 *Critical factors/Scores Mean Score 5.1 6.1 4.6 3.4* 4.7 4.8 4.3 3.7 Mean Score 1.2 2.7 3.8* 2.4 Mean Score 3.0 2.3 2.3 3.6 4.4* 202 The number of berths in the entire Lagos port System is: a) Adequate for present traffic demand :1 3 0) Not adequate for present traffic demand ,5 l:] 11 c) Adequate for present demand but not adequate for the future E 2 The hunter of Gantry Cranes and Fork lifts assigned to a berth are: a) More than adequate at present [3 -- b) Adequate at present D 3 c) Not adequate [:3 13 The capacity of the receiving warehouses (transit sheds) are: a) More than adequate for the present cargo traffic D -- b) Just adequate [:1 10 c) Less than adequate D 6 Cargo handling and stacking in the warehouses (transit sheds) can best be described as: a) Orderly D 5 b) Average E 8 c) Disorderly E 4 Import cargo clearing process in the port takes an average of: a) 1 day D -- b) 2 days ‘ D 1 c) 3 days I: 3 d) 4 days [:1 4 e) 5 days E: 3 f) 5 days [:1 -- g) 7 days [:3 1 h) 8-14 days [:1 3 i) More than 14 days D 1 The present custom cargo clearing process is: a) Sinple and does not need a change 6) Couplicated and needs a change 10 DD 11. 12. 13. 14. 15. 16. 17. 203 The number of supervisory and clerical staff in a warehouse (Transit shed) are: a) More than adequate [:3 -- b) Adequate 1:) 7 c) Less than adequate 1:] 8 Cargo movement out of the port is delayed-due to the fact that insufficient railroad cars are allocated to port operations: a) True ‘ [::j 9 6) False [::l 3 c) Other [:1 2 Containerized iuport goods move through the port faster than general import cargo: a) Yes [::] 14 b) No [::J 1 Containers are returned to the owners: a) On time [::] 2 b) Late ' [::j 9 c) Abandoned within the port area [:Z] 4 Cargo damage in the port is principally caused by one of the following: a) Pilferage (and theft) 12 l b) Poor storage c) Bad weather 1 d) Handling DUDE Longshoremen or dockers should be handled best by: D a) Contractors b) Nigerian Port Authority [:Z] 7 The Tin Can port extension reduced ship and cargo delay by: a) 100% b) 80% c) 60% d) 40: e) 20% f) 10% DDDDDD NHUU'Ib I8. 19. 20. 204 What other major problems do you encounter in the day-to-day operations of the port subsystem (i.e. your Department or Section): a) Lack of adequate operational facilities b) Poor managementgpolicy c) Frustration resulting from too much work and too poor a salary d) Shortage of wagons for delivery of heayy cargo What future problems do you anticipate in Lagos port complex: a) Low staff turnover b) Trade dispute due to poor compensation c) Re-occurrence of major port congestion d) Desertion of experienced staff due to lack of motivation and reward. (4) (2) (2) (2) (5) (6) (5) (4) How do you rate internal communication systems (i.e. radio, telephones, etc.) within the port complex: a) Very efficient b) Efficient c) Good d) Fair e) Poor UDDDD 205 Table (6.14-2). ANALYSIS OF TRAFFIC AND LOGISTICAL OPERATIONS SURVEY FORMS. Category 8: Port Operations Officers. 1. Rank the operational adequacy of the following port subsystems. Assessing values (See KEY) to any of the subsystems order. Mean 1 2 3 4 5 5 7 8 Score a) Signal System B -- -- 1 -- 2 1 7 z 5.5 b) Pilotage and Tugs [:L-- -- 1 I -- 3 4 4 6.5 c) Berths D s 3 3 1 .. 2 1 -- 2.93 a) Gantry Cranes and Forklifts [j 2 3 1 3 1 3 I .- 3.8 e) Labor Gangs per berth G -- -- 7 2 3 1 l. 1 4-3 f) Warehouses (Transit sheds) E 1 2 4 5 1 2 -- -- 3.5 9) Cargo clearing process E I 3 -- 1 7 -- 1 .- 3.3 h) Yard transfer equipment [:1 1 1 1 1 4 3 -- 2 4.9 KEY: Excellent - 8 Good ' 6 Average - 4 Poor ' 2 Very Good - 7 Above Average - 5 Below Average - 3 Bad ' I 2. Rank the major causes of ship delay in the port. _ Mean 0 l 2 3 4 5 Score a) Poor signal, tugs and pilotage [: 4 1 4 2 1 1 1.9 0) Lack of sufficient nunber 0f berths __ I: -- 2 2 I 2 7 3.8 c) Lack of sufficient nutter of Gantry Cranes and Forklifts (loading and unloading) I: -- 1 1 > 3 5 3 3.6' d) Insufficient number of receiving ware- houses (Transit sheds) I: 2 2 1 3 4 2 2.8 e) Labor problems I: z 5 3 2 1 1 1.7 NB Rank should run from 1 t0 5 (i.e. major cause - 5, minor cause - 1) ‘ Major cause - 5. Minor cause - 1. Not as a Factor - O 3. Rank the factors which cause cargo delay in the port of Lagos. Mean 0 1 2 3 4 5 5“" a) Poor handling equipment (Cranes, fork- lifts. conveyors [:1 -- 7 1 I 2 2 2.3 0) Lack of container transfer equipment _ D -- 1 8 3 -- -- 2.2 c) Poor record keeping in transit sheds _ E] 1 3 1 4 2 3 2.9 d) Comlicated custom clearing process __ D 1 3 2 2 S 2 2.9 e) Shipper's unwillingness to clear goods on time [:1 -- 1 -- 2 4 8 4.2. NB Ranks should run fun 1 to 5 (major factor I 5. minor factor I 1) Major factor - 5, minor factors - I, No factor - O 4. Signals, tugs and pilotage services in the port can best be described as: a) Excellent D 3 b) Good I: 7 c) Average I: 3 d) 900'“ D “ *Critical factors/scores 10. 206 The nunber of berths in the entire Lagos port System is: a) Adequate for present traffic demand [:1 2 b) Not adequate for present traffic demand [:1 12 c) Adequate for present demand but not adequate for the future E] -- The number of Gantry Cranes and Fork lifts assigend to a berth are: a) More than adequate at present [:3 4 b) Adequate at present [:3 1 c) Not adequate [:1 IO The capacity of the receiving warehouses (transit sheds) are: a) More than adequate for the present cargo traffic __ D 3 b) Just adequate B 1 c) Less than adequate I: 11 Cargo handling and stacking in the warehouses (transit sheds) can best be described as: a) Orderly D 4 b) Average I 1:] 7 c) Disorderly [:1 4 Import cargo clearing process in the port takes an average of: a) 1 day [::1 -- b) 2 days E -- c) 3 days [:1 4 d) 4 days E 4 e) 5 days [3 2 f) 6 days [:1 I g) 7 days [:3 1 h) 8-14 days [3 2 i) More than 14 days E “ The present custom cargo clearing process is: a) Simple and does not need a change DD 0) Complicated and needs a change 11 11. 12. 13. 14. 15. 16. 17. 207 The number of supervisory and clerical staff in a warehouse (Transit shed) are: a) More than adequate b) Adequate c) Less than adequate 1:] Cargo movement out of the port is delayed due to the fact that insufficient railroad cars are allocated to port operations: a) True [:Z] b) False [Z] c) Other [::] Containerized import goods move through the port import cargo: E] 6 5 faster than general a) Yes [:Z] 0) No L::] Containers are returned to the owners: a) On time , [:1 0) Late [:1 c) Abandoned within the port area 1:] Cargo damage in the port is principally caused by one of the following: a) Pilferage (and theft) b) Poor storage c) Bad weather d) Handling a) Contractors b) Nigerian Port Authority [:1 [:1 [II [:1 Longshoremen or dockers should be handled best by: [:1 [:1 The Tin Can port extension reduced ship and cargo delay by: a) 100% b) 803 c) 60% d) 40% e) 20% f) 10% DDDDDD 12 6 2 2 4 10 NONE.» 18. 19. 20. 208 What other major problems do you encounter in the day-to-day operations of the port subsystem (i.e. your Department or Section): a) Non clearance of cargo by consignees b) Lack of mglti-modal transfer terminals c) Low and irregular plant availablility in thegport d) Low moral among labor because of poor wageslsalaries e) Delays in processing import/export_pagers What future problems do you anticipate in Lagos port complex: a) Labor unrest among,dogkworkers 0) Poor berth occupancy ratio, i.e. 96% c) Congestion of transit warehouses d) The port may not be ablefitg;meet 1990 demand How do you rate internal communication systems (i.e. radio, telephones, etc.) within the port complex: a) Very efficient b).Efficient c) Good d) Fair e) Poor DDDDD U'INNNN (3) (2) (2) (2) (2) (4) (2) (3) (1) 209 Table (6.14-3). ANALYSIS OF TRAFFIC AND LOGISTICAL OPERATIONS SURVEY FORMS. CategorLC: ShipLers. 1. Rank the operational adequacy of the following port subsystems. Assessing values (See KEY) to any of the subsystems order. Mean 123456785core 3) Signal System a -- -- -- 2 2 2 2 5 6.5 b) Pilotage and Tugs a -- -- 2 -- 2 3 5 1 5.9 c) Berths D 7 3 2 .. -- .- .. -. 1.6' d) Gantry Cranes C] -- 3 z 2 -- -- -- -- 2.9 e) Labor Gangs per berth a -- l 1 5 2 2 2 -- 4.7 f) Warehouses (Transit sheds) E] 3 2 7 1 -- 1 2 -- 3.7 g) Cargo clearing process D 3 3 -- 2 4 -- -- -- 3.0 h) Yard transfer equipment CZ] -- 1 2 3 3 l -- 2 4.8 KEY: Excellent - 8 Good - 6 Average - 4 Poor . 2 - 3 1 Very Good - 7 Above Average - 5 Below Average Bad I 2. Rank the major causes of ship delay in the port. Mean 0 1 2 3 4 5 Score a) Poor signal, tugs and pilotage [j 3 3 2 I 2 2 2.2 b) Lack of sufficient nunber of berths _ [:1 1 2 1 2 2 5 3.3 c) Lack of sufficient umber of Gantry Cranes and Forklifts (loading and unloading) D -- -- 3 s 1 4 3.53 d) Insufficient number of receiving ware- houses (Transit sheds) C] -- -- 3 4 7 1 3.4 e) Labor problems [:3 -- 6 2 1 2 I 2.2 N8 Rank should run from 1 to 5 (i.e. major cause - 5, minor cause a 1) KEY: Major cause . 5, minor cause - 1. Not as a factor - O) 3. Rank the factors which cause cargo delay in the port of Lagos. Mean 0 1 2 3 4 5 Score a) Poor handling equipment (Cranes, fork- lifts. conveyors) [j -- 3 2 4 3 I 2 8 0) Lack of container transfer equipment _ 1:] 1 1 6 1 2 2 2.6 c) Poor record keeping in transit sheds __ D 2 2 2 3 -- 4 2.7 d) Complicated custom clearing process _ a -- 2 I 3 6 2 3.6 e) Shipper's unwillingness to clear goods ontime D -- 6 1 I 2 l 2.2 NB Ranks should run from 1 to 5 (major factor - 5, minor factor - 1) KEY: Major factor 8 5, Minor factor - 1, No factor - O) 4. Signals, tugs and pilotage services in the port can best be described as: a) Excellent [3 2 b) Good :1 5 c) Average D 3 d) Poor a 3 *Critical factors/scores 10. 210 The number of berths in the entire Lagos port Systems is: a) Adequate for present traffic demand E] 5 b) Not adequate for present traffic demand D 6 c) Adequate for present demand but not adequate for the future ' :1 4 The number of Gantry Cranes and Fork lifts assigned to a berth are: a) More than adequate at present E 2 b) Adequate at present a 3 c) Not adequate [:Z] 10 The capacity of the receiving warehouses (transit sheds) are: a) More than adequate for the present cargo traffic I: 3 b) Just adequate D 5 c) Less than adequate [:1 7 Cargo handling and stacking in the warehouses (transit sheds) can best be described as: a) Orderly [Z] 0) Average [:1 4 [:1 c) Disorderly Import cargo clearing process in the port takes an average of: a) 1 day b) 2 days c) 3 days d) 4 days uhNe—l e) 5 days H f) 6 days 9) 7 days h) 8-14 days DDDDDDDDD i) More than 14 days The present custom cargo clearing process is: a) Simple and does not need a change 10 DD b) Complicated and needs a change 11. 12. 13. 14. 15. 16. 17. 211 The number of supervisory and clerical staff in awarehouse (Transit shed) are: a) More than adequate D 1 b) Adequate D 10 c) Less than adequate ' D 5 Cargo movement out of the port is delayed due to the fact that insufficient railroad cars are allocated to port operations: following: a) True [:1 4 6) False E] 9 c) Other E] 2 Containerized import goods move through the port faster than general import cargo: a) Yes D 5 b) No [:1 10 Containers are returned to the owners: a) On time D 7 b) Late D 8 c) Abandoned within the port area [Z] I Cargo damage in the port is principally caused by one of the a) Pilferage (and theft) 5 6 0) Poor storage D 4 c) Bad weather B 2 d) Handling [:1 3 Longshoremen or dockers should be handled best by: 3) Contractors I: 7 b) Nigerian Port Authority [:3 6 The Tin Can port extension reduced ship cargo delay by: a) 100% D -- b) 80% [:1 -- c) 50% [:1 3 d) 40% [:1 1 e) 20% :1 4 f) 10:; E 2 18. 19. 20. 21. 2112 What other major problems do you encounter in the day-to-day operations of the port subsystem (i.e. you Department or Section): a) Poor shipper information 0) Difficgty of cargo tracing c) _fllgh user gharges d) e) What future problems do you anticipate in Lagos port complex: a) Intolerable cargo delay by 1980 6) Excessive pilferagle and cargo delay c) High berth occupancy d) Increase in ggrgo insurance rates (5) (4) (3) (5) (6) (4) (3) How do you rate internal coumunication systems (i.e. radio, telephone, etc.) within the port complex. a) Very efficient b) Efficient c) Good d) Fair e) Poor Cl [:1 Are you prepared to direct your shipments to the ports of Warri and Koko if these ports offer quicker cargo clearning facilities than Lagos. Yes [:Z] I No [::J 16 REFERENCES Spivey, Allen and M. Thrall: Linear Optimization, p. 164. Dantzig, George 8.: Linear Programming_and Extensions. Princeton University Press, Princeton, N.J., 1963. Gass, Saul: Linear Programming Methods and Applications. McGraw- Hill Book Company, New York. Hadley, G.: Linear Programming. Addison-Welsey Publishing Company, Inc., Reading, Mass., 1964. Luenberg, D. G.: Introduction to Linear and Nonlinear Programming. Addison-Welsey Publishing Company, Inc., Reading, Mass., 1973. Lemke, C. E.: "The Dual Method for Solving the Linear Programming Problem." Naval Research Logistics Quarterly, Vol. 1, No. 1, 1954. 213 CHAPTER VII ALTERNATIVES AND RECOMMENDATIONS Three major alternatives have been identified for detailed evaluation: (i) Do nothing (ii) Development of the ports of Warri, Koko, Burutu and Port Harcourt; with the hope that these ports will significant- ly divert ship and cargo traffic from the port of Lagos. (iii) Improvement of the quality of service offered by the port of Lagos by implementation of the results of this research. In selecting these alternatives the author was searching for long term solutions to the problems of the port of Lagos. Heavy investment in floating cranes and utilization of roll-on roll-off ships may provide temporary reduction in ship and cargo delay (1). Alternatives in this category were not considered. The Do Nothing alternative is not a functional solution to a major national problem (2). Average ship waiting time in the queue during 1978 was 215 hours (i.e. 9 days) while service time at berth was 199 hours (i.e. 8.3 days). This means that the total delay in the port system averages 416 hours (i.e. 17.5 days) including transit time. The demurage and service costs associated with these delays are: Demurage cost per ship =$105 x 215 = $22,575 Berthing cost per ship =$200 x 199 = $39,800 Total cost of ship service $62,375 214 215 Cumulative annual cost = $62, 375 X 5000 $311,875,000. where $105.00 the demurage cost per ship hour $200.00 5000 the berthing cost per ship hour the number of ships which entered the port of Lagos in 1978. As a result, a do nothing alternative will not be in the interest of national economy. This annual service cost represents 1% of the GNP (3) of Nigeria. When other costs of port congestion such as the lag in the industrial and construction sectors are considered the social cost of port congestion can be tremendous. Port congestion reduces employment opportunities by creating a downward multiplier on economic activities. In developing countries inflationary pressures can be brought about by the increased cost of imported goods. Hence a 'Do nothing alternative' is not in the interest of balanced economic growth and development. Alternative (ii) should be considered in light of the spatial distribution of Nigeria's ocean ports. Table (7-1) illustrates the relative distance between Lagos and any other ports. Port Harcourt the country's second largest general cargo port is 315 nautical miles from Lagos. This means 13 hours of sailing time for a ship making a maximum speed of 24 knots. Hence the ports of Port Harcourt and Lagos do not serve the same geographic market. This means that even if Port Harcourt offers a better quality of service, Lagos based shippers will not consign their cargo through that port. The third largest general cargo port is located at Calabar which is even further than Port Harcourt (394 nautical 216 Table (7-1) . TABLE OF DISTANCES PORT TO PORT (NAUTICAL MILES) 'o 3: c J a .4 . o g 8 . . in g h > U a ... e .9 a >~ E I O In 5 3 a 5 3 = a 33 d3 13 a: 3i :3 13 .( to c: :3 c: <9 LAGOS 126 153 158 184 171 192 221 288 330 315 315 394 °Eacravos Lt. Ho. 36 4O 46 48 69 119 186 228 213 213 292 'Forcados 5 27 59 80 149 216 258 243 243 322 'Burutu 32 64 85 154 221 263 248 248 327 'Warri 86 107 175 242 284 269 269 348 “Koko. 21 167 234 276 261 261 340 “Sapele 188 255 297 282 ; 282 361 Akassa 83 1 25 1 09 1 09 1 90 Bonny 42 27 57 136 Degema A 69 99 178 Port Harcourt 84 163 Opobo 101 Calabar “Distances assume a crossing of Escravos Bar when a seaward voyage is undertaken. Source: Nigerian Port Authority: Handbook 1976. Nigerian Port Authority 26/28 Marina, Lagos. 217 miles). The same argument as in Port Harcourt applies to the diver- sion of Lagos traffic to Calabar. Apart from the ports of Warri and Koko the rest of the ports in figure (7-1) are specialized bulk ports. The ports of Warri and Koko are the nearest general cargo ports lying 184 and 171 nautical miles from Lagos respectively. Even these ports involve an addi- tional 7-8 hours of sailing time from Lagos. This implies increased freight costs for Lagos-based shippers. It is obvious that shippers will be unwilling to absorb this additional cost of cargo diversion. A number of geographical and operational inadequacies also dis- qualify Warri and Koko ports as alternates for the port of Lagos: - Warri and Koko ports are inland river ports with an approach channel that is subject to seasonal draught variations. At high water the draught rises to 25', and drops to 15' at low water (4). Hence year round navigation for all classes of general cargo ships is not possible at reasonable cost. - The channel length of 35 nautical miles (5) from the Atlantic coast to Warri and Koko would require an annual dredging cost of over $16 million. - Warri port has a berthing capacity for six general cargo ships while Koko port has four berths (7). These facilities are grossly inadequate to receive ships diverted from the port of Lagos. 218 - The port of Warri and Koko are not integrated into the national rail network system. This means that movement of heavy and bulk cargo will be difficult. As shown in figure (7-1) a total of 150 miles of rail track will be required to connect these ports to Ibadan, which would require a capital outlay of $50,000,000 (8).f' . The results of the logistical operations survey indicate that only one out of every sixteen Lagos-based shippers would be willing to consign cargo through the ports of Warri, Koko and Port Harcourt. In general, the author is of the opinion that the ports of Warri and Koko should not be considered as alternates to the port of Lagos. These ports serve a different geographical market, and deserve development of their own but not as alternates to the port of Lagos. Alternative (iii) involves the implementation of results of the Simulation and the optimization programs. As shown in tables (6.7) and (6.11) case 3 is the most viable combination of port resources for servicing short term demand while case 6 applies to optimum long term port investment requirements. Figure (7-1). Source: 219 RAIL NETWORK INTEGRATION OF THE PORTS 0F WARRI AND KOKO NATIONAL r. I .l .’ )- I. .. i\ I 2 '\ o l. I {I < I.’ O . 4’ I / P Yola (21) 'h’ddnmrgeuml . 4??o,.< 3 9 (J) on Nger I J‘PP‘ . Bar-o gr“ .I' g “0) ('19) u to d. .J O 1 ’Ubadan ,mfinaw I s 1, L01“)! Mekurdi ' l I ”v r ’8'. ./ a , "7: flats: r" , ,J Q 511%: Epe ‘,,. " j ‘0 ...; {are ,‘ .I' \J S“ LAGOS a nitsha rang ' ‘59”I. c ' ./."(' F. °e : ‘arri ” O V .s ' . I bl f 80:44.13 PORT. g“ ! $$ \ 3130.73): gravy/it 84 M of Benin HancouRT gum ‘5 steamer: etc. .9 M \PDON c p (’0’ Numbe'o’mcrchs A 093‘ wearer-W114: this: : 9“ ... Ravi s a , Right of Slob-e W W. A. Perkins and J. Stembridge: Nigeria: A Descriptive Geography. London University Press, Ibadan, 1962. 220 As shown in table (7-2) the cost reductions which will be achieved if the results of this research are implemented are compared with the present operational costs of the port of Lagos. The implementation of the optimum combination of investments as determined from case 3 in table (6.8) results in an annual cost savings of $191,224,000.00. This figure represents a 61% reduction when compared with the status quo or do nothing. The above cost savings is the aggregate result of the addition of 5 new berths and extra annual investment in equipment, labor, warehouses, and other logistical subsystem defined in Chapter 3. As shown by the optimization results, the implementation of the above alternative requires an annual investment of $194,025,418. This figure appears high but when compared with the tangible savings of $191,224,000 a benefit to cost ratio of .99 is obtained. There are also intangible benefits associated with an efficient port operation; particularly in a developing country where the economy is highly import dependent for producer goods. An upward multiplier on the economy is created by increased cargo thoroughput. This situa- tion results in an increase in employment due to higher levels of industrial, agricultural, construction and transportation activities. When these secondary impacts are considered the benefit to cost ratio obtained above will be understating the economic value of total benefits to the nation. Hence the simulation and optimization alternative should be implemented. The details showing the optimum level of improvement required in logistical subsystems are specified in tables (6—8) and (6-11). 221 .mmmfi Low ooom u z .mea ago so ugoo xvsum on“ ocwgmucm mowgm mo Logan: m>wumpzszo an umou Emu? mg» ocwapopupze An nocmmuno umou pmzcc< Pmuoha compuoomc m mmmo RS .m..p 8.53.2. . a a a a a a . IPHQO Ucm ooo omm HoH ooo Hmo oNH ooo ooo o- ooN oNH ooo Hmo moH «N H comampzswm E: ooo mm_umcgmp_< owzm oo goo umoo mswp momgzsmo we va Loo; me_h mocp>om we umoo “moo _mzcc< mswp ozone ocwcugmo «moo Fmacc< goo umoo ocpu_m3 umoo —m:::< Poacc< pouch «Pouch -gmo owgm owzm «pouch mongoewo mawzo m>_umcgmup< .mu>~pmhommmm hmou .AN.~V mpaoh 222 The results of the logistical operations survey indicate that par- ticular attention should be focused on the following co—ordination efforts: ~ Data recording system at the port of Lagos should be revised to generate a more comprehensive set for demand forecasting and delay calculations. The present system of ship and cargo delay recording does not specify idle time at various port subsystems. - The two independent departments of Port Statistics and Traffic Statistics should be merged to ensure concentration of efforts. - Further research is required in the area of the cargo clearing process and custom requirements. Computerization can speed up the rate of processing cargo and ships. ° Communication within the port system needs additional investment as indicated by the logistical operations survey. This will facilitate cargo tracing and clearing. ° Regular monitoring of ship and cargo traffic through the port will generate reliable information for the management decision process. Note that the logistical co-ordination efforts have been stated in broad terms because this research concentrates on traffic analysis, simulation and optimization of physical subsystems. The logistical operations survey was introduced to help in identification of major coordination problems. These problems require in depth study in order to determine specific solutions. The custom cargo clearing process is an example of such a problem. The present process involves a time lag 223 due to the tremendous amount of paper work involved. Shipper information is another area which requires an in depth study. Daily publications should be established to provide information on ship schedules, routes and cargo location. The results of the survey also indicate that the internal communication systems (telephones and radios) within the port are far from adequate. Detailed investigation should be carried out to assess the need and level of improvement required. The importance of logistical co-ordination efforts cannot be over— emphasized. In the author's opiniom maximum thoroughput and efficiency cannot be achieved unless considerable co—ordination efforts are ensured. The statement above presupposes that an optimum combination of physical subsystems exists. It is also important to update port subsystems in response to changing demand. This process requires accurate data re- cording and reliable forecasts. When this modification is not carried out the probability of failure is higher and the associated congestion cost can be tremendous. Hence port development should not be considered on an ad hoc basis, but on a long term basis. Continuous traffic and operational survelliance is required in order to identify changing demand and service variables. These changes provide inputs to executive decision process. REFERENCES "Port Congestion in Nigeria": a paper presented to the Fourth Conference of the Port Management Association of West and Central Africa. August 2-6, 1976. Nigerian Port Authority 26/28 Marina, Lagos. Nigerian Ports Authority: NPA News SpecialgiTin-Can Island Port Edition. Nigerian Port Authority 26/28 Marina, OctOber 1977. His Excellency General Obasanjo: National Budget Speech 1978-79. Federal Ministry of Information, Lagos, Nigeria. Nigerian Geographical Society: Location Factors in Changing Sea- port Significance in Nigeria, 1970. Ibid. .Nigerian Port Authority: NPA News, op cit. Nigerian Port Authority: Handbook 1976. Nigerian Port Authority 26/28 Marina, Lagos. W. A. Perkins and J. Stembridge: Nigeria: A Descriptive Geography. London University Press, Ibadan, 1962. Bowersox, Donald J.: Logistical Management. New York: Macmillan Publishing Co., Inc., 1974. 224 APPENDICES APPENDIX A RESEARCH DATA COLLECTION FORMS 225 mongoum mucom cmoo acmEumm>:_ ocmo Hg mmzosmgoz uwmcmgh 0H mocmo AmuwWP xcoov acmso_=ou ocwpocm: mcmco zgucmo l\ Ammgzuuogamv gucmo oogmo _mcmcmo ooh omapopwa o:_oomco mmwupppumu mongoguc< NMQ'LDND Emumxm .mco_m memEmm gmosoz Pouch Amcmmxv mow; mow>cmm “moo apnacza> rmzzc< umou mucmcwucwmz rm:=:< “moo _mowaao Emamzmoom pmuwumpooo mo msmz .oz Sago zzooxz~ Paco .Afiso.vv mpaop 226 pcmso_:ou :owumugoomcmgh ocmpcm>o mo mmvuwppomo cowamugoo -mcmc» czoogm: Luca“ «H ucwsopocu comment» ucm> ma mumom omumgmowgoma NH Amgmqu umoo umoo consaz mow; mpnw_cm> mucmcmu:_mz umoo smumzmnom .oz mgcmswm pouch muw>cmm poocc< szcc< _mumomo —muwummooo oo mamz swam Atmocwucouv zzoo¥zm Ema; 227 . ozaco 395 25.6 39.... 3...: we: 3...: we: Page ..3 L33 ..8 Page .5”. 253 ..8 .2532. :5 o: 3:53 9.58:5 9:» a: o: a a o: a a: 3.5:: 7...: 3...: 6.3.3:“ 323 2953: 35 .223:— pouoh ~33 9:: £33 2.: Eono—c: 895» 332:. 9.2.2: 75mm; :2: as: 9:25 2 9:: mzupm>moom boom 3:255: oz: «8 35:52: 38 25. .Awuoév oz: 228 Roosmv Acmmx Amcou xuppoo Lmov ooh ooh gucmm cu mozh oo moo—mo u_cumzv mup>gmm we “moo oo “moo mongogoco soc; gmosoz mo zuwumomo ooh Ewan mxm, rmumomo we?» muw>cmm Page». cmofioz ocwzop oo mozh .oz mm4m<~m<> ww<30h oz< mwsum vousaomo ouao cacao :. us_p oc—u_a: sagas»: ouao guocud “amass: mmugoxu someommao .xpao .oaasp .ooguu pacacoo o.gm .u._ .o~.m ecu mosh no use: .oz mm4m<_¢<> o—zm .Aeio.ev mpaoh ZKBO assuaoz a».uoaau unsogocu: o:.>—ouo¢ mu -cooauuo gran «sop oc.uaop -== 3.3m ooaco>< oa_p ou_>com o=.uao4 a.gm ooagm>< mun—woo; sommcach go=_au=oo mono a a. cue -ucogmocoo gonna: muc.o stag ua goons: mucosa acacao oa Loans: guano .oz auaoma canon mmam¢~¢<> :hzum .Am-o.e. «_aa_ 231 meeEem mseumam commons» gexe>eeo Locum xgem_> ugmozm me sensoz emee; -egez cw eeceu gene; Axum: Leno emceepo eoecee» eoceo Pouch Axum: Leov ee>weeem eoecceh eogeo Peach Aecw_ xcoco ageuee>:_ acesepecm Au» eweeev emee; nose: we zapeeoeo .ez e_eeo emeezegez AHHmZ<¢Po meem ezemaezem<3 .Ae-e.eo m_aah 232 A.mgso Aegee; eesmepo mueo ee>weeem eueo A.mggv me_uppveeu Le mzeov ececo zgaeeo geomeegh eoeceum Ameeeem ceee twee: eswh excesem Ameeuv uzo_e3 —ecgeu:_ cm eewp ecu emeezegezv oepuwez eogeo .ez so each me_ew_wemm meageem getem seem Amzo_hoz:m mzmh moo mmmhmz<4mo owm< dNH'IfiOBQO «nah; 4<_¢mm m—zm -\nn~s.so anxamsn.uo ceamcsm.ao «sxavoa.os ocmuaqn.cs ¢~=nnog.cu nounnvc.ms .«ss-8.eo wuaccoa.no Joanc-.co waoeoaa.vm nmuanva.mm an~=«o~.ov nnouman.mv raavmou.¢c cazuuau.¢v .nn=a-.cn coaacNa.mn «~smmv=.c~ economa.ca avas~e~.c« enunmo~.m« moaanom.na oon¢ona.a« woavann.o wvasodo.c wnaaoon.c novnosv.m .122. . uz_p 4<>_¢¢< ~2n2n85.vmd «mamoco.cvu ooaaofi«.¢.« anamsgo.coa vcmnncq.n¢« mongono.cnu muaan‘v.snd onnus‘v.vn« osvmo¢«.un« on~mqga.enu ansencu.eda m-.nmo.nau o¢nmosn.mea «nm.o«°.co ecavomv.ec samumxn.qe «nesman.sn manevmn.soh mmuuaeu.ce nvomcce.=o soocs‘~.en maovosu.fin «mnas‘s.en enahn¢o.d~ ooo..n«.o« seamo¢«.nd «sousnu.sa magmnuv.ma Agcg. u:_p uu_>¢um SERVICE TIME (hnL) 157.0767269 158.7695672 159.27327;a 164.2966051 168.5578490 171.0631452 100.0520814 147.0707572 187.5024946 203.20 8920 205.7956436 211.4539599 221.8549753 226.2080715 234.0750520 235.75206£3 237.1342095 244.40062°3 246.2047192 256.1:83657 259.7‘62311 263.9305493 267.3875622 275.6553859 207.335cELO 297.2991449 300.77910/4 301.824:043 311.8‘110.7 313.1052155 328.8553567 333.26794/8 335.5747472 339.1911479 341.6795377 352.5“21495 352.9°077*7 365.7277475 369.35351i9 314.9"0¢9°1 367.8730415 399.824:724 412.240¢855 414.4550727 417.6051954 427.9765152 456.7621955 430.0140277 430.30727‘6 437.2675851 440.7006699 440.12018é0 444.1010774 450.2699139 462.2085757 469.6618192 471.9674176 476.5278531 476.26964c3 484.9708279 63°.7P7¢2*1 ARRIVAL TIME (hnL) 105.1699115 106.0030400 107.3611133 108.4324953 11142355423 112.0156240 112.3664923 115.9060965 110.8370269 119.0491577 120.6201070 123.2663116 1236‘955999 130.3149006 130.935592‘ 133.2266135 134.9460/54 137.7418846 133016396‘0 141.0320913 144.4920127 146.4362060 149.4685960 155.4523906 155.5766917 161.0347113 164.5186011 165.060‘,3, 16660705520 1660“afl.9fl 1710223856“ 173.4214180 177.164152? 18061192515 166.3508703 190.4800930 190.8753‘43 195.0117557 19°01‘7095‘ 20102540606 210.2633983 214.2370140 21848890936 224.0640963 220.6745‘69 230.9112‘13 232.1092663 235.5507’6i 236.1504026 236.4059030 230.5614624 230.0021220 238.9676790 246.1933’74 240.9512147 257.0963979 2606n16127‘ 260.7904644 263.6100112 265.2940920 26505967105 247 SHIP SERIAL mnmsn 29 30 31 32 33 34 35 36 37 36 39 AVERAGE BERTHING “HE , 5(.h4r1334§3h6‘g ) 5.2923323 5.1370475 5.1530190 5.1070136 5.0312690 5.1672023 5.2061321 5.0676350 5.6495496 5.2766369 5.9363490 5.4110970 5.3859065 5.9436059 5.6560014 5.2692046 5.6365354 5.6022261 5.6357993 5.3009435 5.2796049 5.2420934 5.3010652 9.4214310 5.5036730 5.4667110 5.6097196 5.4706965 5.3904175 5.5730196 5.5544656 5.5012254 5.4706250 5.4234647 5.5092523 5.4306274 5.5413295 5.5127391 5.6616219 5.7655600 5.7116367 5.9062491 5.7563204 5.7206192 5.7834719 5.7301653 5.6500793 5.5491067 5.6059947 5.5709602 5.5391023 5.4637169 5.4910965 5.6049275 5.5935930 5.5525579 5.5410216 5.5000466 5.5110321 5.5023351 QUE LENGTH 2521330000 23.0000000 22.0000000 23.0000000 25.0000000 24.0000000 27.0000000 27.0000000 26.0000000 30.0000000 2°.0000000 30.0000000 30.0000000 30.0000000 32.0000000 32.0000000 33.0000000 35.0000000 36.0000000 35.0000000 35.0000000 37.0000000 33.0000000 42.0000000 43.0000000 52.0000000 53.0000000 51.0000000 53.0000000 53.0000000 56.0000000 5°.0000000 61.0000000 61.0000000 62.0000000 66.0000000 65.0000000 66.3000000 66.0000000 72.0000000 74.0000000 74.0000000 60.0000000 79.0000000 7°.0000000 86.0000000 65.0000600 64.0000000 63.0000000 63.0000000 64.0000000 83.0000000 63.0000000 87.9000000 69.0000000 00.0000000 69.0000000 00.0000000 67.0000000 86.0000000 65.0000000 SERVICE TIME (hum) 494.4717415 504.2016750 507.6973922 510.2455034 511.0135920 513.6541976 513.6692470 513.9‘3e850 517.7551234 521.7007079 526.7574756 52°.1I‘9C5-fio 531.6302839 539.9466057 547.2442420 550.2052231 556.}F54236 556.&?§97.9 560.9794133 563.19553.5 588.2916133 5°1623e27“ 592.39665’4 600.4470595 601.1?17991 601.2’491‘2 601.4r314~5 605.91'5IE’33 610.55612‘3 619.0‘32172 6216237283”) 621.5565219 622.3155572 623.0560091 625.36126T6 625.6i21055 534.7039836 640.7597642 653.2‘77653 655.2241150 559.7612614 662.5:659s1 662.7361041 666.5721214 671.9649233 676.7779367 636.9?19037 695.6:57163 706.2946975 707.4309337 710.2?05727 712.2050473 716.5634525 716.6045646 722.4365695 724.6650427 726.3“65152 727.53041‘1 746.6449596 740.6:56434 751.20142i1 ARRIVAL TIME (hrs.) 200.0121066 266.9106’10 26963680403 270.4106‘56 27".”801181 275.8155234 207.1657136 269.1747367 28'6577799’ 289.1937014 291.742?61h 291.875021i 292.1005137 293.1666‘05 295.2474uon 296.7666944 297.1314699 299.0391030 303.5460566 311.4704975 311.2466967 312.6614055 315.9349545 316.‘139992 325.2625175 326.820365? 330.0630554 330.9705355 331.4913050 332.1502477 333.3335‘23 3336535679, 334.1914306 330.0775959 340.1637913 340.9209019 343.?061‘06 344.2575950 346.3694937 3506572903: 352.3835930 3530501012. 359.6976365 35‘61375/4; 359.1667837 367.6540‘1f 375.791412? 376.9029614 379.7516805 361.1740520 38300783999 339.1156910 390.1250’65 391,407,520 399.4295075 ‘026236’11‘ 404.5397053 405.4910773 407.3309010 400.4404fl2q 410.5714577 248 SHIP SERIAL MfiflER 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 10° 110 111 112 113 11‘ 115 116 117 118 119 120 121 122 123 124 125 126 127 126 129 131 132 133 134 135 136 137 136 139 140 141 142 143 144 145 146 147 146 146 150 AVERAGE BERTHING RATE (pr5./ship) 5.4941305 5.5406777 5.5196455 5.4065151 5.4448297 5.4068862 5.3507276 5.2903665 5.2632155 5.2697041 5.2675747 5.2367094 5.2136264 5.2422000 5.2619639 5.2400783 5.2300512 5.2320465 5.1942529 5.1668434 5.3472674 5.3265475 5.2004702 5.3136906 5.2729979 5.2282150 5.1646623 5.1765110 5.1742231 5.2020438 5.1769774 5.1368341 5.1009497 5.0718424 5.0432406 5.0054566 5.0373333 5.0453525 5.1034982 5.0791792 5.0750866 5.0573969 5.0207432 5.0118220 5.0146636 9.0131999 5.0656023 5.0774070 5.1174253 5.0892673 5.0730069 5.0516670 5.0462215 5.0256263 5.0167817 4.9990693 4.9753214 4.9491665 5.0564119 5.0256093 5.0120095 QUE LENGTH (ships) 67.0000006 66.0000000 65.0000000 84.0000000 63.0000000 63.0000000 82.0000000 61.0000000 61.0000000 62.6006066 64.0000000 63.0006000 63.0000000 63.0000000 66.0006060 66.0000006 87.0000000 67.0000000 66.0000000 65.0000000 99.0060000 90.0000006 97.0000000 99.0000060 90.0000000 97.0000000 90.0000000 90.0000006 9°.0000000 99.0000000 90.0000000 97.0000006 96.0000060 95.0000000 94.0000000 93.0000000 92.0000000 93.0000066 94.0000006 94.0000006 95.0000000 94.0000000 93.0000060 92.0000000 93.0000000 94.0000000 94.0006006 90.0000000 101.0000000 101.0000066 101.0000000 101.0000000 100.0000066 99.0000000 99.0000066 99.9006066 99.0000006 90.0000006 104.0000006 103.0000006 103.0000000 APPENDIX c SYNOPSIS OF UNCTAD PORT SIMULATION MODEL 249 SYNOPSIS OF UNCTAD PORT SIMULATION MODEL FUNCTIONS OF PORT SIMULATION PROGRAMS AND ITS SATELLITE PROGRAMS CRITERIA: The principal criterion in a port simulation model is time requirements in the entire port system both for ship and cargo. Time is a major cost function in port operations. PORT SIMULATION PROGRAM COMPONENTS: The UNCTAD port simulation model can be broken up into five phases: Satellite Programs (A) (i) Date Accumulation Program (ii) Forecasting Program (iii) Traffic Generator Program (B) (iv) Main Simulation Program (v) Output Program The five phases of the simulation program enable greater flexi- bility to be achieved. The split programs also enable computers with less than 32K words core memory to handle each phase of the simulation rather than one large program. The attached flow chart illustrates the UNCTAD Port Simulation model. The various programs are briefly discussed below: (1) DATA ACCUMULATION PROGRAM: This program reads all the punched cards containing data related to ship characteristics (e.g.. arrival time. type, amount and nature of cargo). After reading such informa- tion the program produces a frequency list. This frequency list is coded as (FRELIS). 250 Plan of the simulation programme READ DATA l ACTIVITIES - Activity SHIP (which in addition to the body of the actiirity contains procedures PILOTLEAVES, TUGLEAV ES. BERTHTIME, CALTIM. TEST. BERDAT. SORT) - Activity SEASON - Activity SHIFT - Activity HIGH — Activity PILOT - Activity TOWAGE I r—H HOLD ONE YEAR A . l 1 TEST CONVERGENCE l PRINT CUMULATED TIMES FOR SHIPS AND CARGO Source: United Nations: Improvement of Port Opera-. tions for Seaport Growth and Development. MIT, July 1969. 251 FRELIS contains eight different types of information on ships and cargo: (a) Ship type (e.g.. liners, tramp, bulk. etc.) (b) Ship tonnage (gross registered tons) (c) Type of import carried by ships (e.g., specialized cargo. refrigerated cargo, not specialized) (d) Type of export carried by ship (categorized as in c) (e) Volume of export cargo carried by the ship (in tonnage QrOUPS) (f) Volume of import cargo carried by the ship (in tonnage groups) (9) Fraction of bulk cargo carried by ship in relation to total cargo (exports) (h) Fraction of bulk cargo carried by ship in relation to total cargo (imports) (2) FORECASTING PROGRAM The Data Accumulation Program prints out a frequency list based on the present port situation. The forecasting program creates a new frequency list for future port situations. This new frequency list or FRELIS is based on the present port conditions and trade forecasts of ship and cargo traffic through the port system. It is important at this stage to note that the forecasting program does not predict future ship and cargo traffic trends. These forecasts are obtained from trade statistics of the economic port environment. The forecast- ing program above is limited to the reproduction of a future frequency list given future ship and cargo flow forecast. The new FRELIS pre- dicts what new information on ships and cargo are likely to occur in the future [see items (a). . . .(h) in the Data Accumulation Program above]. 252 (3) THE TRAFFIC GENERATOR The traffic generator translates the data contained in the fre- quency lists into a traffic pattern (present or future). In situations where ports have seasonal variations a traffic pattern will be required for each major season. This implies that a new frequency list for each season will be obtained from the forecasting program mentioned above. In summary, the traffic generator performs the following functions: * Determines the arrival times for ships on the basis of time between them. I * It predicts the interarrival time by making use of observed ship arrival distribution and applying average time intervals. * This program also incorporates ship and cargo priority policies into the frequency list (i.e. FRELIS). * The program records all the generated parameters for each ship and repeats the entire process for the next arrival. The traffic generator creates a traffic pattern which provides an input to the main Simulation Program. (4) SIMULATION PROGRAM The main simulation program is made up of a master program and several activities. The master program controls the sequence of the activities. Each activity contains several procedures for handling specific problems. The input to the program are as follows: * traffic pattern obtained from the traffic generator * list of technical standards assiged to the various port sub- systems (e.g., number of pilots, tugs, cargo handling area, stations. queing area, out-of-port transportation) 253 The program (SIMULA) sets up a time axis. At time '0' the program activities start with each activity marking a new event on the time axis. All events are simulated in sequence of occurrence. In order to highlight the design the various activities and procedures are briefly discussed (see Flow Chart II). 1. Procedure PITUGALL: Allocates pilots and tugs when necessary. 2. Procedure Nounites: Determines number of loading and unloading units available. 3. Activity Ships: Controls ship's movement in the port system. 3.1 Procedure PILOTLEAVES AND TUGLEAVES: separates the ship from pilot and tugs when necessary. 3.2 Procedure CALTIM: records cumulative time for different cargo and ships for all seasons. 3.3 Procedure Berthtime: computes unloading and loading time for each ship. 3.4 Procedure Test: considers alternative loading and un- loading possibilities if the first unit is not acceptable. 3.5 Procedure BERDAT: indicates the % of berth occupancy and identifies vacant berth. 3.6 Procedure SORT: queues ships and sorts them according to priority policy of the seaport. 4. ACTIVITY SEASON: makes provision for seasonal traffic and physical variations. 5. ACTIVITY SHIFT: keeps record of the time left of the present shift and determines the number of equipment available for the next shift. 6. ACTIVITY HIGH: Records the water depth and tides in various 254 sections of the port. 7. ACTIVITY PILOT AND TONAGE: Keeps account of the number of pilots and tugs in the system and matches these with ship when necessary. It is important to note that all the activities and procedures mentioned above are for a generalized port situation. In specific ports there may not be need for some of the activities. After the simulation of one-year operations the program performs a test of con- vergence. The cumulative time consumed by a ship in the simulated nth year is compared to the time consumed by a similar ship in (n + 1)th year. This is only true when successive years have similar traffic flow patterns. Nhen adequate convergence is established the computer prints out the results. The printout contains the time requirements for the entire port and also specific ship and cargo time at various port subsystems. APPENDIX D BACKGROUND OF THE AUTHOR RESUME OF SAMUEL KINGSLEY NNAMA HOME ADDRESS OFFICE ADDRESS No. 1533 L Spartan Village Department of Civil Engineering East Lansing, MI 48823 Michigan State University Telephone: (517) 355-2919 East Lansing, MI 48824 PERSONAL Date of Birth: June 14th, 1945 Height: 6'5" Place of Birth: Awka, Nigeria Height: 175 lbs. Marital Status: Married Health: Excellent CAREER OBJECTIVE Permanent: Pursue a consulting career in civil engineering and areas related to transportation planning and logistics distribu- tion systems (design and management). EDUCATION Ph.D. Michigan State University, East Lansing. Michigan. Major Civil Engineering, Transportation and Highway Engineering M.S. Michigan State University, East Lansing, Michigan. Graduation date: June 1977 Major Transportation Planning and Highway Engineering 8.5. University of Nigeria, Nsukka. Nigeria. Graduation date: June 1973 (2nd class honours upper division). Major Civil Engineering EMPLOYMENT EXPERIENCE Summer 1977 - June 1979 Division of Engineering Research, Faculty of Engineering. Michigan State University. Title: ‘Ggaduate Research Assistant (level 2) Project: Development and Presentation of Short Courses in Transporta- tiOn Engineering -- (a) Traffic operations course. (b) Highway capacity course. (c) Highway safety course. Participants include State, County and Local traffic engineers in the State of Michigan. The above course is jointly sponsored by The Federal Highway Administration. 255 256 Supervisors: Professors James Brogan and William C. Taylor Winter 1978 - June 1979 Department of Civil Technology, Lansing Community College. Lansing, Michigan. Title: Part Time Instructor. Courses: Civil engineering and related courses. Winter 1977 - Spring 1977 Civil Engineering Department. Michigan State University, East Lansing, Michigan. Title: Teaching Assistant Courses: CE 347 Transportation facilities CE 499 Highway engineering Supervisor: Professor James Brogan PROFESSIONAL EXPERIENCE January 1975 - September 1976 I worked as an executive engineer with the Federal Ministry of Works. Lagos, Nigeria. December 1974 - July 1973 I was an executive planning engineer with Kano State Ministry of Works in Nigeria. June 1966 - January 1968 Served as trainee engineer with NNAMA Shipping Lines (Nigerian Limited). PROFESSIONAL ASSOCIATIONS (i) Member of the Nigerian Society of Engineers (ii) Member of the American Society of Transportation Engineers (iii) Registered Engineer in Nigeria HOBBIES Music. Art, Tennis BIBLIOGRAPHY 10. 11. 12. 13. 14. 15. BIBLIOGRAPHY - Agerschous. H. and Korsgaard, J.: Systems Analysis for Port Planning. The Dock and Harbour Authority, March 1969. . Ahrenholz, 0. J.: Network Analysis of a Ugter Port Cpmplex. Trans- portation Network Analysis System and Terminal. Proceedings of Meeting, April 17-18, 1969; October 1969. . Bennington, G. and Lubore: Resource Allocation for Transportation, Naval Research Logistics Quarterly, Vol. 17, No. 4, December 1970. . Bowersox, D. J.: Logistical Management. Macmillan and Co., London, 1974. . Chenery, Hollis B.: The Application of Investment Criteria, Quarterly Journal of Economics, No. 47, February 1953. . Collier, P. 1.: A Simulation Model for Ports Management Training. Dock and Harbor Authority, Harcourt Street, LondOn NIH 2AX, England. . Computer Services for Port and Waterway Management. Dock and Harbour Authority, London. V52 N616. February 1972, pp. 431-432. . Cooper, Robert 8.: Introduction to Queueing Theory. Macmillan, New York, 1972. . Creton, Jean-Michel: Review of Optimizatipn Methods Applicable to a Computer Aided Design of a Queueing System Using a Simulation and Stochastic Processes. CTDL, MIT, 1973. Cruon, R.: Queueinngheory_(Recent Develppments and Applications). American Elsevier, New York, 1967. Da Silva, F. M.: Boletin do Porto de Lisbon. No. 150, 1963. Dantzig, George 3.: Linear Programming and Extensions. Princeton University Press, Princeton, N.J., 1963. Dierexens, H. 3.: Impact of Containerization on Integrated Distribution. Economisch en SoCiaal Tijdehnift 25 (5), October 1971. Drew, Donald R.: Traffic Flow Theory and Control. McGraw-Hill, New York. Economics Associates: Nigerian Eorts: Traffic and Development. Vols. I and II, London, 1967. 257 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 258 Embassy of Nigeria: Federal Nigeria_(1976-1977). 2201 M Street, N. W., Washington, D.C. 20037. Ericksen, 5.: Optimum Cppacity of Ships and Port Terminals. Department of Naval Architecture and Marine Engineering, University of Michigan, Ann Arbor. Ericksen, Stian: Simulation of Receiving,p§toring and Loading General Cargo. University of Michigan, Department of Naval Architecture and Marine Engineering, Ann Arbor, Michigan. Federation of Nigeria: Nigeria Trade Summagy (1970-1977). Federal Office of Statistics, Lagos. Federation of Nigeria: Review of External Trade. Federal Ministery of Statistics, Lagos (1970-1977). Ferris, C. D., F. E. Grubbs and C. L. Weaver: Operating Characteristics for the Common Statistical Tests of Significance. Ann. Math. Statistics. Frankel, E., P. Wilmes and K. Chelst: Simulationpf Multipurpose Port and Multiport Offshore Facilities. Offshore Technology Conference, 6200 North Central Expressway, Dallas, Texas 75206. Fratar, T. J., A. S. Goodman and A. E. Brant: Prediction of Maximum Berth Occupancy. Journal of the Waterway and Harbours Division. Proceedings of the ASCE, 1960. Fulkerson, Delbert R. and George B. Dantzig: Computation of Maximal Flows in Networks. Naval Research Logistics Quarterly, Vol. 2, No. 4, December 1955. Garfinkel, Robert and George Nemhauser: Integer Programming. Wiley, New York, 1972. Gass, Saul: Linear Prpgramming Methods and Application. McGraw-Hill Book Company, New York. Glover, J. W. and H. C. Carver: Introduction to Mathematical Statistics. Edwards, Ann Arbor, Michigan, 1928. Goldman, A. J. and A. W. Tucker: "Theory of Linear Programming." Paper 4 in Kuhn and Tucker, Annals of Mathematics Studies 38, 1956. Gooneratne, S. G. and D. J. Buckely: Operations Research Models for Bulk Handling_Systems at Sea Transport Terminals w1th Particular Reference to Port Kembla. Report N. 2, School of Traffic Engineer- ifig, NSW, 1970. Goss, R. 0.: "Towards an Economic Appraisal of Port Investments." Journal of Transport Economics and Policy; London School of Eco- nomics and Political Science; Houghton Street, Aldwyck, London WC2A 2AE, England. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 259 Gould, F. J.: A Linear Programmin ng Model for Cargo Movement Evaluation, Transportation Science, Vol. 5, No. 4, 1971. Gramer, H.: Mathematical Methods of Statistics. Princeton University Press, Princeton, New Jersey, 1946. Greenberg, H.: Integer Programming. Academic Press, Inc., New York, 1971. Guy, Arnold: Modern Nigeria. Lowe and Brydone, Ltd., Norfolk, England, 1977. Hadley, G.: Linear Prpgramming. Addison-Wesley Publishing Company, Inc., Reading, Mass., 1964. Hadley, G.: Nonlinear and Dynamic Programming. Addison-Wesley, Reading, Mass., 1964. Haight, F. A: Index to the Distribution of Mathematical Statistics. Journal Of Nat. Resource, Bureau Standards Section 865,1961. Haley, K. B.: A General Method of Solution for Special Structure Linear Prpgrams. Operational Research Quarterly, Vol. 17, No. 1, 1966. ‘ Hansen, J. B: Optimisinngorts Through Computer Simulatipn, Sensi- tivitygAnalysiSL of PertinentT Parameters. Operations Research Quarterly, Vol.23,No. 4,1972. Harrison, J. 0., Jr.: "Linear Programming and Operations Research." Operations Research for Management, Vol. 1, John Hopkins, Baltimore, Maryland, 1954. Hazard, John L.: Tran§portation Management Economics Policy. Cornell Maritime Press, Inc., Cambridge, Maryland, 1977. Heulbroner, Robert L.: Understanding Macro-economics. Prentice-Hall, Inc., London, 1975. Hirschfeld, D. 0.: Mathematical Programming Development at Management Science Systems, Inc. Presented to SHARE XXVIII, March 9, 1962. Hobbs and Moore: Financial Accounting Concepts, Evaluation, Analysis. Sourth-Western Publishing Co., Cincinnati, 1974. Hu, T. C.: Multi-commodity Network Flows. Operations Research, Vol. 11, No. 3, 1964. Hu, T. C.: Integer and Network Flows. Addison-Wesley, Reading, Mass., 1969. Hyvarinen, Lassi: Mathematical Modeling for Industrial Processes. Springer Verlag, Berlin, New York. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 260 Integer Programming Methods for a Vessel Scheduling Problem. Transpor- tation Science, Vol.5, No. 1. 1971. Jaiswal, N. K.: Priority Queues. Academic Press, New York, 1972. Johnson, K. M. and H. C. Carnett: The Economics of Containerization. London: Allen and Unwin, 1971. Klaasen, L. H. and N. Vanhove: Macro Economic Evaluation of Port Investments. Paper presented at Bruges Week, College of Europe Semaine de Bruges, 1970. Kuhn, H. W. and A. W. Tucker: "Linear Inequalities and Related Systems." Annals of Mathematics Studies 38, Princeton University Press, Princeton, New Jersey, 1966. Lasdon, Leon 5.: Optimization Theory for Large Systems. Macmillan, New York, 1970. Lee, A. M.: Applying Queueing_Theory. Macmillan, London, 1966. Lemke, C. E.: "The Dual Method for Sovling the Linear Programming Problem." Naval Research Logistics Quarterly, Vol. 1, No. 1, 1954. Lloyd, 0. and M. Lipow: Reliability, Management, Methods and Mathe- matics. Prentice-Hall, Englewood Cliffs, New Jersey, 1964. Luenberger, D. G: Introduction to Linear and JNonlinear Programming. Addison- -Wesley Publishing Company, Inc. , Reading, Mass. , 1973. Mahmoud, M. H.: On the Problem of Demurage. Arab Maritime Research Journal, Arab Maritime Transport Academy, Alexandria, Egypt, 1975. Manetsch and Park: Systems Analysis and Simulation withpApplication to Economic and Social Systems. Mettam, J. D.: Forecasting Delay to Ships in Port. The Dock and Harbour Authority, VOl. XLVII, No. 558, 1967. Mettam, J. 0.: Berth Planninggpy Evaluation of Congestion and Costs. Journal Of the Waterways and Harbour Division, Proceeding ASCE, 1968. Meyer, J. R., M. J. Peck, J. Stenason and C. Zwick: The Economics of Competition in the Transportation Industries. Harvard University Press, 1959. Miller, A. J.: Queuing at a Single Berth ShippingTerminal. Journal of the Waterways and Harbours Division, Proceedings of the ASCE, 1971. Mills, G.: "Investment Planning for British Ports." Journal of Economics and Policy, London School of Economics and Political Science, May 1971. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 81. 261 . Molina, E. C.: Poisson's Exponential Binomial Limit. Van Nostrand- Reinhold, Princeton, New Jersey, 1947. Morse: Queue, Inventories and Maintenance. John Wiley. Mossman, Frank, Paul Bankit and Helferich: Logistics Systems Analysis. University Press of America, Washington, D. C. 20023. Nagorski, B.: Port Problems in Developing Countries. Dock and Harbour Authority, Harcourt Street, London. Neumann, J. Von: A Numerical Method to Detenmine Optimum Strategy. Naval Research Logistics Quarterly, Vol. 1, No. 2, 1954. Newell, G. F.: Application of Queueing Theory. Chapman and Hall, London, 1971. Nicoloan, Stairso N.: Berth Planning_py Evaluation of CongestiOn and Cost. Journal of the Waterways and Harbour DiVision, Proceedings ASCE, 1967. Nigerian Geographical Society: Location_Factors in Changing Seaport Significance in Nigeria. (1970-1977). Nigerian Ports Authority: Yearbooks (1970-1977), NPA. 28 Marina, Lagos. Nigerian Produce Marketing Company: Summary of Annual Returns of Monthly Shipments. Lagos (1970-1977). Nigerian Statistical Yearbook (1970-1977). Federal Ministry of Statistics, Lagos. Orner, Ron: Port Simulation Program. MRIS Publication, Washington, D.C. Parsons, Ron and Lawrence Hill: Analysis and Simulation of a Seaport. Commodity Transportation and Economic Development Laboratory, MIT, January 1973. Perkins, W. A. and Jasper Stembridge: Nigeria, A Descriptive Geography, Univeristy Press, London, 1962. Port Development in the United States. Maritime Transportation Research Board; 2101 Constitution Avenue, Washington, D.C., January 1976. . Robinson, R.:v Sequential Linkages and Spatial Structure: Intra-port Shipping Movementand Port Development Policies. 4th ANZAAS Congress Perth, 1973. Robinson, R. and K. P. Tognetti: The Structure of Shipping_Inputs to the Port of Port Kembla. Report No. 2 ARGC Project Wollongong University College. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 262 Robinson, R. and K. P. Tognetti: QueuingpModels and_the Operational Structure of Ports. The Port of Port Kembla. Report No. 3. ARGC Wollongong University College. Robinson, Ross and Keith Tognetti: Modelling_and Port Policy Decisions: The Interface of Simulation and Practice. Marine Services Board. New South Wales, Australia. Rochdale Report: Committee of Inquiry_into International Shipping. HMS Commd 4337, London. Rossa, G.: Investigation of Ship Arrivals in a Line Service. Seewirt- schaft, Berlin, East Germany. Vol. N 11, November 1969. Schenker, E.: Future General Cargo Traffic and Terminal Rgguirements at the Port of Milwaukee. Center for Great Lakes Studies, University of Wisconsin, Milwaukee, 1968. Scott, R. Pearson: Measurement of the Impact of Petroleum Production on the Nigerian Economy. (Mimeographed), 1969. Simmonard, Michel: Linear Progpamming. Prentice-Hall, Englewood Cliffs, N.J., 1966. Spivey, Allen and Robert Thrall: Linear Optimization. Holt and Winston, Inc., New York. Taborga, Pedro N.: Determination of an Optimal Policy for Seaport Growth and Develppment. MIT, July 1969. Tomlin, J. A.: A Mathematical Prpgpamming Model for the Combined Distribution Assignemnt of Traffic. Transportation Science, Vol. 5, No. 2, 1971. Trace, K.: Underdeveloped Countries: Shipping Problems and Policies. The World Today, May 1968. United Nations: Yearbook of International Trade Statistics (1950-1970). United Nations: Improvement of Port Operations for Seaport Growth and Development. MIT, July 1969. United Nations: Appraisal of Port Investments. TD/B/C.4/174 UNCTAD, Geneva, 1977. White, Schmidt and Bennett: Analysis of Queuing Systems. Academic Press, New York. Whittington, M.: Pre-Investment in Port Facilities. Dock and Harbour Authority, Harcourt Street, London W1H 2AX, England. 263 98. Williams, Phillipe: Port Analysis and Simulation. Commodity Transpor- tation and Economic Development Laboratory, MIT, 1972. 99- Wilmes, P. and E. Frankel: Port Analysis and Plannipg, MIT. 100. Wohl and Martin: Traffic Systems Analysis for Engineers and Planners. McGraw-Hill, New York.