-. - -1 .,_' "4-,, 9‘s.- 1 17‘}? a" ‘5' ; 1 I 1f} ,fifil‘gffi #154”?! 1 "3;“? .1. fl . 3‘ ~91»; . d4 r, (if ‘fi Afé- ",1 A 3; 95' ' .. J a V “‘17, (a; : ‘u :3; as" ‘ ‘“ *3»- 35:65? 2‘.’K:A‘;".'-'§. \ .. 13,3; , ‘ ,w,‘ . “1. ‘ 4' . ”51.116. , .‘ 53h “2,3 . g“ r. A I‘ v \ i ,3. 9 «Hess 1“ ANSTATEU 1111111111111111111111111111111111111 23 01024 0103 :11111211 This is to certify that the thesis entitled A STUDY OF PACKAGE DYNAMICS IN SMALL PARCEL ENVIRONMENT OF UNITED PARCEL SERVICE AND FEDERAL EXPRESS presented by AMRITPAL SINGH CHEEMA has been accepted towards fulfillment of the requirements for MASTER degree in PACKAGING M r professor DateM 0.7639 MS U i: an Affirmative Action/Equal Obportunity Institution LIBRARY Michigan state University PLACE N RETURN BOX to roman this checkout from your record. TO AVOID FINES mum on or Moro duo duo. ‘ DATE DUE DATE DUE DATE DUE fin, , _"u_: 7 ,, 112411711 51 £300 _‘ W03 ’f MSU In An mm Wand Opportunity trunnion Wm: W 1A STUDY OF PACKAGE DYNAMICS IN SMALL PARCEL ENVIRONMENT OF UNITED PARCEL SERVICE AND FEDERAL EXPRESS BY Amritpal Singh Cheema .A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1995 ABSTRACT A.STUDY 0F PACKAGE DYNAMICS IN SMALL PARCEL ENVIRONMENT OF UNITED PARCEL SERVICE AND FEDERAL EXPRESS BY Amritpal Singh Cheema The purpose of this study was to measure and characterize free fall drops, kicks and tosses occurring in the overnight small parcel environment of United Parcel Service and Federal Express. Five Drop Height Recorders (DHR) were repeatedly sent through UPS "Next Day Air" and Federal Express "Priority" services, respectively, to five destinations in the U.S. from East Lansing. These destinations were Monterey (CA), Atlanta (GA), Rochester (NY), Portland (OR), and Memphis (TN). In a total of 50 roundtrips, 2394 impact events were recorded. The data showed that a package encounters 24 shock events on an average one-way trip which consists of 31% drops, 43.6% kicks, and 25.4% tosses. The highest free fall drop height was from 77.8 inches. The maximum kick level was 233 in./sec, and highest equivalent drop height in a toss was 31.4 inches. Of all the drops, 95% were from less than 16 inches drop height, 95% of the kicks were from less than 135 in./sec, and 95% of tosses were from less than 10.5 inches. equivalent drop height. The packages received 51.1% of impacts on edges, 42% on corners. Only 6.9% of all impacts occurred on flat faces. Dedicated to my Parents iii .ACKNOWLEDGEMENT I express my deep sense of gratitude to my major advisor Dr. S. Paul Singh for his able guidance, constructive criticism and useful suggestions during the course of the present investigations. I also express my sincere thanks to Dr. Gary Burgess for his expert guidance and help rendered in conducting this study. My thanks are also due to Dr. Galen Brown for his input and critique of the manuscript. My thanks are also due to my friends who supported me in various ways in carrying out the work undertaken. Finally, I am grateful to my parents for the inspiration and warm affection which helped me to focus on my work. iv TABLE OF CONTENTS LIST OF TABLES ........................ . ......... ... ....... vi LIST OF FIGURES .......................................... vii 1.0 INTRODUCTION.. ........................................ 1 1.1 Next Day Air System ..... ... ............. . ........ 3 1.2 Objectives ......... ... ....... . ................... 9 2.0 EXPERIMENTAL DESIGN ............. . .................... 10 2.1 Test Instrumentation and Packages ............... 10 2.2 Drop Height Recorder Operation.. ................ 13 2.3 Zero-G Drop Height Calculation .................. 16 2.4 Equivalent Drop Height Claculation .............. 16 2.5 Instrument Setup Parameters ..................... 18 2.6 Instrument Configuration and Calibration ........ 23 2.7 Data Collection. ...... ......... ...... ...........26 3.0 DATA.AND RESULTS ............. . ..... . .......... .......31 4.0 CONCLUSIONS ........................................ ..58 TEST PROTOCOL ............................................. 60 TEST SEQUENCE.. ............... ..... ................ .......61 LIST OF REFERENCES ................ . ..................... ..62 APPENDIX .................................................. 65 LIST OF TABLES Table Page 1. Summary of.A11 Drops........... ..................... .32 2. Summary of.A11 Kicks ................................. 34 3. Summary of All Tosses. ...... .... ............... ......36 4. Cumulative Data for.All Shipments.. ............... ...38 5. Summary of Impacts for.All Shipments ................. 41 6. Cumulative Percent as a Function of Impact A1 Level for All Shipments .............................. 43 Frequency of Total Impacts as a Function of Orientation.... ......... . ....... . .................... 51 Frequency of Drops, Kicks, and Tosses as a Function of Orientation... .......... . .............. ..53 Distribution of Impact Orientation as a Function of Drops, Kicks, and Tosses.................56 Individual Drop Events in Overnight Small Parcel Environment of UPS and Federal Express ............... 65 Individual Kick Events in Overnight Small Parcel Environment of UPS and Federal Express ............... 82 Individual Toss Events in Overnight Small Parcel Environment of UPS and Federal Express ........ . ..... 106 vi LIST OF FIGURES Figure Page 1. Drop Height Recorder Tri-axial Orientation ........... 12 2. Block Diagram of DHR Operation............. .......... l4 3. .Map of Overnight Destinations in the U.S....... ...... 27 4. Flow Path of UPS and Federal Express 10. 11. 12. 13. 14. Package Delivery System............... ......... ......29 Distribution of Impacts for All Shipments... ....... ..40 NUmber of Drops versus Drop Height for All Shipments.44 Cumulative Percent versus Drop Height for Drops in All Shipments...............................45 Number of Kicks vs Velocity Change for.A11 Shipments.46 Cumulative Percent versus Velocity Change for Kicks in All Shipments........................ ....... 47 Number of Tosses versus Equivalent Drop Height for.A11 Shipments............................... ..... 48 Cumulative Percent of Equivalent Drop Height for.A11 Shipments.............. ..... .......... ....... 49 Frequency of Impacts as a Function of Package Orientation ..... . ...... . ............................. 52 Distribution of Impact Type as a Function of Package Orientation .................................. 54 Distribution of Package Orientation as a Function of Impact Type ....................................... 57 vii 1.0 INTRODUCTION Packaged goods have been moved from.one place to another using various forms of transportation. There has been a continuous increase in the number of goods that are shipped and handled. The distribution environments that packaged goods are subjected to are continuously improved to handle and deliver more packages with increasing efficiencies for time and cost. Several of these emerging distribution systems therefore consist of complex networks. These distribution systems have been operated by both Federal as well as private companies. These carrier companies employ various modes of transportation such as trucks, rail, air, and water. United Parcel Service (UPS) and Federal Express are two of the major private U.S. companies offering a one class service for door to door shipment of small parcel packages. These companies handle millions of packages every day. UPS is a major force in the flow of commercial goods, both on the ground and in the air. Its development has spanned.much of the modern transportation age, beginning early in this century. The UPS was originally established in 1907 under the name of American Messenger Company (UPS,1993). Over the years the company merged and evolved into UPS in 1919. In the early 1950's UPS decided to acquire common carrier rights 2 to compete with U.S. Postal Service, a Federal agency. In 1987, UPS became the first package delivery company to provide service to every address in the U.S. The air delivery system. was started in 1929 when UPS opened United Air Express, offering package delivery along the west coast. This service was terminated in 1931, then restarted in 1953. By 1980, "UPS 2mi Day Air" service was started and in 1982 entered the overnight air delivery business with "UPS Next Day Air” service. By 1985, overnight air delivery was available to every address in the U.S. and now includes more than 185 other countries (UPS, 1993). UPS has a fleet of 197 jet aircraft and 260 chartered aircraft covering a total of 961 domestic and 536 international daily flight segments. Today, UPS has an average daily air delivery volume of 875,000 packages (UPS Next Day Air and 2nd Day.Air combined). This accounts for about 8% of a total daily delivery volume of 11.5 million parcels and documents handled by this company. The bulk of the packages are delivered by ground transport (UPS, 1993). Federal Express first started its business in 1973 from Memphis, TN. Federal Express has grown since then and has specialized only in air delivery systems. In 1989, with the acquisition of Tiger International Co. and the integration of Flying Tigers Co. into its system, Federal Express became the worlds largest full service all—cargo international airline. 3 Federal Express now provides overnight air deliveries to virtually every address in the U.S. through its next day ”Priority" and ”Standard" services (Federal Express, 1993). The "Priority" service guarantees delivery of a parcel by 10 a.m. the next working day, whereas the "Standard" service ensures that the packages will be delivered by 5 p.m. the next day. Federal Express handles an average daily package volume of 1.7 million parcels in its combined air delivery services. A.tota1 of 465 aircraft are used to make daily connections to 186 countries worldwide. 1.1 NEXT DAY AIR SYSTEM Both. UPS "Next Day .Air" and Federal Express "Priority” services use the "Hub-and-Spoke" system to deliver packages. The local operating centers all around the U.S. serve as the "spokes". Each operating center provides pickup and delivery service within an individual territory. The all-cargo aircraft connect these local operating' centers with. the Central Air-Hub. The "Hub" is a single central sorting facility. The aircraft called "Feeders" take a consignment of packages from the local operating centers to the Central Air- Hub for sorting every night. These packages are sorted at the Air-Hub in a matter of approximately three hours. After sorting the packages, the aircraft depart with a load of 4 packages to be delivered the next morning at the destination operating center. The UPS operates 2250 local service centers while Federal Express has approximately 1400 such operating centers worldwide (UPS, 1993). While the concept of picking up and delivering documents and packages is simple, systems that make it operate efficiently and reliably are innovative, complicated, and expensive. The task of sorting over a million. packages at one facility, and. placing these in containers for return shipments in a matter of hours, without errors is a continuous challenge. During the distribution of packaged goods, damage during handling and sorting is inevitable. Once the packaged product is shipped through a distribution system such as of UPS or Federal Express, it is subjected to a series of hazards such as drops, impacts, crushing forces, vibration, climatic, and pressure changes, before it reaches the customer. All manufacturing, engineering, and quality efforts are in vain if the product reaches its destination in a damaged condition. The factors that contribute to the damage of a product during handling and distribution are numerous. Shock is one of the more severe and commonly occurring hazards in the small parcel shipping environment. Shock occurs when a moving package comes in contact with a stationary object, either a package or a surface. Shocks often result from.packages being dropped, tossed, and sorted. 5 All these can occur as packages are handled and sorted manually or by automatic sorting equipment. Many studies have been done to uncover primary features of shock and vibration that relate to product damage during transportation. The distribution environment may be mainly categorized into handling environment and in-transit environment. The damage in a handling environment generally results from operations such as loading-unloading, stacking, lifting, and conveying packages that occur in sorting and storage areas. On the other hand, damage during in-transit environment result from transport on vehicles (trucks, railcars, aircraft, etc.). The severity of the damage varies with distance as well as surface of travel. Most of the previous studies have investigated the dynamic characteristics of the in-transit environment. Some of the recent studies are reviewed in this section. Hausch (1975) studied vibration and its interaction with package systems in the transportation environment. The study showed that the truck vibration environment seems to be the most severe at low frequencies. The rail environment showed severe shocks resulting from switching of boxcars. The aircraft shipping environment had higher acceleration levels at higher frequencies when compared to truck and rail. In another study conducted by Marcondes (1988), the 6 dynamics of three different package types were studied in a Less-Than—Truckload (LTL) shipment. The study showed that accelerations as high as 10 6'5 (16 = 386.4 in./sec2) were encountered during vibration in packages at the top of the stack. Packages with low natural frequency show more bouncing and larger acceleration levels than those with higher natural frequencies. Singh et a1. (1992) compared the lateral and longitudinal vibration levels with vertical vibration levels in the truck distribution environment over various highway conditions. Power Density Spectrums were developed for various road conditions. The study showed that lateral and longitudinal vibrations above 20 Hz were similar to vertical vibrations, but were very low at frequencies below 20 Hz. Pierce et al.(l992) studied the effect of suspension types in the ride qualities in trailers. The results showed that the air ride suspension produced lower vibration levels on all road conditions examined as compared to leaf spring suspension. Also, the damaged air ride suspension systems showed similar response frequencies to the leaf-spring suspension systems but caused higher acceleration levels. There have been fewer studies that have monitored the dynamics of individual packages as they are handled and sorted. Ostrem and Godshall (1979) compiled an assessment of the common carrier shipping environment. The major shipping hazards of shock, vibration, impact, temperature, and humidity 7 associated.with the handling, transportation, and warehousing operations of typical distribution cycles were documented. The loads imposed during handling operations have been reported in terms of drop heights. The study reported the occurrence of a large number of low level drops, and very few drops with higher levels. The study concluded that heavier and larger packages were dropped from smaller heights. .Also most packages got dropped on their bases representing over half of all drops experienced by the package. The drop height data in the Ostrem and Godshall study was collected by several methods including observation, camera, and instrumented package. Some of these methods could result in large errors. Data on several loading conditions on the cargo deck of the C- 5A.and C-141 aircraft during normal operations like run-up, takeoffs, cruise, landings, taxi and extended flights was also analyzed. It was reported that worst conditions occurred while flying in turbulent air and during landing. Voss (1991) measured the dynamics of the small parcel environment in the UPS ground shipping environment. The effect of weight and size was also studied. The study used packages of different sizes and weights that were instrumented with drop height recorders. The results showed that the highest drop height measured was 42.1 inches. The size of the package had no significant effect on drop heights. Weight did not have a significant effect on the medium and larger size 8 packages. However small size lighter weight packages experienced higher drop heights. This was attributed to more automated handling for the larger and heavier packages for the UPS sorting environment. The smaller and lighter packages are often placed on top of the delivery loads and therefore are subject to higher drops. Changes in temperature and pressure during ground-air movements in air shipments may also cause problems to some sensitive products (UPS, 1975). There have been continuous changes in the methods by which small packages are handled and transported over the last decade. There has been a sharp increase in the number of packages handled every day to be delivered next day by companies such as Federal Express and UPS. It is important to characterize the dynamics of the next day air small parcel environment. This information can be used with product fragility information to better design and test packages for this shipping environment. 1.2 This OBJECTIVES study had the following objectives: To measure the dynamics of the next day small parcel environment for Federal Express and UPS in the U.S. Develop a test protocol to test packages for the next day air small parcel environment. 2.0 EXPERIMENTAL DESIGN The goal of this study was to evaluate the small parcel environment of overnight air delivery systems used by United Parcel Service (UPS) and Federal Express. In order to achieve the goals of this study, the test was designed to obtain and collect dynamic data that could be used to develop test methods to simulate this shipping environment. The detailed description of the test instrumentation and packaging used is presented in this chapter. There are several types of instruments used to document and measure dynamic events that packages are subjected to. These range from single drop counters which only record that the package was dropped above a pre-set height, to recorders that measure impacts in the three axes of the package. These data recorders measure and save the acceleration-time history for the dynamic event, and such data can be used to estimate the actual height and orientation of the drop. 2.1 TEST INSTRUMENTATION.AND PACKAGES One such recorder is the "Drop Height Recorder (DHR)" manufactured by Dallas Instruments, Dallas, Texas. It is commercially available to companies to measure the shock environment of a package distribution system. The DHR unit is 10 11 a portable, battery powered, microprocessor controlled, memory device which stores digitized waveforms of shock events sensed by its internal tri-axial accelerometers. Refer to Figure 1 for DHR tri-axial orientation. Test packages were used to package the recorders (DHR) to measure the different events that the packages experienced in the next day small parcel environment. The test packages consisted of 3 components: a DHR, a static shielding bag, and six polyurethane side cushions. All the components were obtained from Dallas Instruments Inc. The cushions give the Drop Height Recorder a particular coefficient of restitution that is subsequently used in calculations. .Although, the DHRs are made of rugged construction able to withstand harsh conditions, the cushions were also meant to safeguard the instrumentation from structural damages or abrasions when exposed to severe shock inputs in the distribution environment. The static shielding bags prevent electrostatic discharge buildups during handling as these may be potentially damaging to the Drop Height Recorders. The DHR units were encased using 1 inch thick polyurethane (open cell) cushions on all sides. The units were placed in the geometric center of double wall corrugated boxes. The cushions provided a snug fit to the recorder. .All DHR's were placed in the same orientation in all packages for every shipment. The packages were closed with a 2 inch wide 12 Fig 1. Drop Height Recorder Tri-axial Orientation 13 general purpose plastic box sealing tape (H-seal) using the 3M automatic case sealer..All the packages were of the same size, weight, and shape. The size and weight of the packages were as follows: Size: DHR Recorder: 6.625 x 6.625 x 6.625 inches Test Package: 10.25 x 10.25 x 10.25 inches Weight: DHR Recorder: 9.5 lb. Test Package with Recorder: 12.5 lb. 2.2 DROP HEIGHT RECORDER OPERATION This section provides a brief review of the DHR operation and also describes the parameters used to program these devices to collect the data. Refer to Figure 2 for a block diagram of DHR operation. The model DHR-1c Drop Height Recorder is a small, light-weight, solid state device that uses a tri-axial accelerometer to record shock events. The DHR constantly monitors the environment and saves events that exceed a predetermined threshold shock level. The recorded events are saved for analysis using software provided with the recorders. The recorder acquires and stores two separate types of data. When a significant impact occurs, the tri- axial accelerometer with a built in pre-amplifier sends the 14 Triaxial Accelerometer (Measures ambient accelerations) 3-Channel Low Gab Amplifier System r Waveform S’ nals 3-Channel High Gah Amplifier System or Zero-G 8‘ na Zero-G Signal Rectification Multi Channel Dblizer Three axis Waveform Shnals 8 Kbyte Digital Delay Trigger Threshold Comparators if any Signal Emeeds 'Trigg r Level' Acquied + Auxiliary Data yte (Temporary Storage) 512 Kbyle Permanent ontro , ispTay 8. Interface To Compute Fig 2. Block Diagram of DHR Operation 15 conditioned signal of an impact to two 3-channel amplifiers. One of these is a low-gain system for sensing and recording high acceleration signals and the second is a high-gain system for recording free fall (zero-G) conditions. The tri-axial accelerometer data is processed by the high gain amplifier system which senses changes to zero-G state (free fall condition) of the unit in a drop. This data is processed as a separate fourth channel called "zero-G channel". The same tri-axial output is also processed by a 3-channel low gain amplifier and is saved as acceleration-time history for the three axis. The information for these four channels is saved in 8 K-bytes of digital delay memory. The peak acceleration is then compared with the threshold trigger level. If the trigger level is exceeded by any of the three low gain amplifier channels the data is diverted to 32 K-bytes of RAM for temporary storage while the CPU processes the data for peak acceleration. The CPU then directs the acceleration data along with information such as time, date, temperature, battery voltage, etc., to 512 K-bytes RAM fbr long term storage. The CPU also stores the key summary data regarding the event in non-volatile RAM along with event pointer and wrap counter. The system is then reset by the CPU to acquire the next record. 16 2.3 ZERO-G DROP HEIGHT CALCULATION The recorder calculates the drop heights using two different methods. The first method is called the Zero-G drop height calculation. The free fall time is measured by sensing the change in the recorder from a motionless state (zero-G), into a free fall (16), and a shock state (several G). Since the time from the onset of the zero-G state of the recorder to the moment of impact is known, the free fall drop height is calculated by following relationship: where, Acceleration due to Gravity = 386.4 in./sec2 "33‘ ll Free fall drop height in inches Free fall time in seconds {'1' II 2.4 EQUIVALENT DROP HEIGHT CALCULATION Equivalent drop height is calculated using impact acceleration time history of the 3-channel digitized shock pulses for each event. Velocity change is first calculated from each acceleration time curve for a given channel. Equivalent drop 17 height is then calculated for each channel using the following equation: 11 = —’.— ................ . - ' (he) 29‘ (22) where, In = Equivalent Drop Height, inches AV = Velocity Change, in./sec. e = Coefficient of Restitution g = 386.4 in./sec2 The coefficient of restitution (e) is determined based on actual drop tests performed during instrument calibration in the laboratory. This is used in subsequent calculations to determine equivalent drop height for the three axes. Finally, the component equivalent drop heights for each of the axes are summed to obtain resultant equivalent drop height as follows: 11 = (h; + by + bx) ..................... (2-3) where, Total equivalent drop height, inches :3‘ a II = Equivalent drop height for each axis, inches x,y,z 18 2.5 INSTRUMENT SETUP PARAMETERS The DHR-1C software supplied with the recorders allows communication between the computer and the units. It is used to configure and calibrate the units before shipment and for uploading' recorded. data from. DHR into the computer for tabulation and analysis. The summary reports that are generated include information such as event number, date, time, battery voltage, temperature, pulse widths, peak acceleration (each axis and resultant), normalized acceleration (%), event frequencies, velocity change, equivalent drop height, zero-G drop height, and deviation between equivalent drop height and zero-G drop height. The configuration is a set of user programmed instructions which must be defined in order to obtain useful data. The configuration allows the user to set operating parameters, trip information, time and date, and alarm settings. Once defined the configuration is downloaded into the DHR. The key configuration parameters include trigger level, memory retention mode, and data acquisition mode. The trigger level defines the minimum.G levels that have to be exceeded for an impact to be recorded by the unit. It is defined in percent of full scale acceleration level for each channel. Each 1% represents 16 of shock, if the full scale is 100 G. Setting a lower trigger level would allow 19 more events to be recorded, rapidly filling up the available memory. The pre- and post-trigger times also need to be set in order to capture the leading and trailing edges of the shock waveform. When the DHR is exposed to a dynamic force which exceeds the trigger level, the resulting' waveforms are recorded that are within a pre-defined window of time. The duration of this window is determined from pre- and post- trigger times set earlier by the user. The size of the window limits the maximum number of events that can be recorded I before filling up the DHR.memory. The memory retention mode settings determine how events are to be stored in the DHR memory. There are essentially three different ways of data storage. In the "FULL/STOP" memory mode the unit will record acceleration events until the bulk memory is full. Once full the unit automatically goes into standby mode without further event recording. This allows only the first set of events exceeding the threshold being saved. In the "WRAP" memory mode, the unit will first record events until the bulk memory is full. The event pointer is then reset to zero and the wrap counter is advanced by one. Any subsequent/new event will overwrite the oldest data in bulk memory. This allows for retaining only the most recent events in case the memory is filled. 20 If the "MAX" memory mode of operation is used, the summary data from all bulk memory events is stored in a non- volatile RAM (memory). The summary data includes peak acceleration values for each axis along with date, time, event number, battery voltage and temperature. Once the primary memory is filled, the peak acceleration of a newly acquired event is compared with the lowest value in the summary data memory. The new event is recorded only if its peak value is greater than lowest peak value of one of the previously recorded events. If the criterion is met, the new waveform data and its summary data replace the lowest data in both the bulk memory and the non-volatile summary data memory. This process of reviewing the event summaries and replacing them with the new event only if it exceeds the lowest values previously recorded, results in saving the most severe events. The data acquisition mode determines how and when the data will be acquired by the DHR. A repeat cycle timer is used to program up to 3 different modes of data acquisition in addition to normal triggered mode (NORM). The timer initiated recording programs the DHR to become active to receive and record any event during a specified window of time regulated by a cycle timer. This allows the DHR to be turned on and off at specified times. The acceleration events occurring only in that time frame are recorded. All events outside of this window of time are disregarded, no matter how severe the shock 21 level. ”NORM? is the default mode of operation in which the DHR is configured to operate in an event-triggered fashion. The DHR is continuously monitoring the environment and records acceleration event data only if one or more of the accelerometers exceed the pre-defined trigger levels. In the "SNAP" mode the unit becomes active at a specified cycle time interval. It takes a snapshot of the environment and acquires one record regardless of signal level. It goes into ”hibernation" again until the next cycle time interval is reached. It is mainly useful for sampling the distribution environment for mainly DC types of signals such as temperature, relative humidity, voltage (pressure), etc. In the "TRIG" mode, the timer-initiated recording of triggering events, the DHR becomes active at a specified time interval regulated by cycle timer. The unit remains active until a triggering event occurs, which is recorded, and the unit goes back to hibernation until the next timer interval is reached. This mode is primarily useful for statistical sampling of triggering events. In the "BOTH" mode, the functions of "NORM" mode and the "SNAP" mode are combined. The unit continuously monitors the environment and only records events that exceed the trigger level. In addition, it also becomes active at a specified time interval and records a single event regardless of signal level and then again goes into normal triggered mode. 22 Once the key operating parameters have been set, the configuration is downloaded into each of the DHR units. The downloading operation deletes all existing data and clears the DHR memory. The event pointer and wrap counter are reset to zero. It also sets the instrument timer to the computer's internal clock. Every time the unit records an acceleration event the time and date of occurrence is also tagged as part of the record. This allows the user to characterize any "hot spots" of the distribution environment and spread of the data over time. The internal electronics of the DHR are powered by two attached battery packs each containing six, series connected, 4.9 Amp-hour, D cell, nickel cadmium.batteries. For the DHRs to function properly, the batteries must be charged to optimum. levels. Fully charged batteries may last up to 14 days depending on the temperature and sampling rate. Operating time is the longest under ambient temperatures (68° F). Any variation from the optimum temperature will reduce battery life in direct proportion to the difference of temperature from the optimum. Higher sampling rates will also decrease operating times. The DHR units were recharged for at least 24 hours before shipment to obtain reliable data. Since a single roundtrip was completed.in approximately 4 days, the batteries retained enough charge to ensure proper functioning of the DHR units. 23 2.6 INSTRUMENT CONFIGURATION AND CALIBRATION. To ensure proper functioning of the instrument, simulated laboratory package impact tests were performed to calibrate the DHRs. For calibration the DHRs were dropped from known free fall drop heights of 18 and 30 inches using a Lansmont Precision Drop Tester (PDT). The recorded data was uploaded into the computer using DHR software to verify that the instrument was accurately recording and measuring the events it was exposed to. Before laboratory calibration, the DHRs were configured with the Operating parameters (variables) that would allow to acquire the most useful and accurate data. The operating variables include the trigger level, pre- and post-trigger samples, sampling rate, memory retention mode, and data acquisition mode described earlier in this chapter. The objective is to set a high enough trigger level that would avoid continuous triggering of the DHR-1c to very low level impacts that do not cause damage to packaged products, and yet be sensitive enough to record a majority of impacts. Based on prior experience a trigger level of 10% (106) of full scale was used for all units. The pre- and post-trigger times to be recorded. were set at 750 and 1000 milliseconds, respectively. The sampling frequency was 1000 samples/sec. The memory retention mode was set to "MAX" to save events with 24 maximum impact. The data acquisition was set to "NORM" to record events that exceeded the trigger threshold. The impacts or shocks observed in the distribution environment may be categorized into free fall vertical drops, lateral tosses, and kicks. .A free fall "drop" may occur when a package falls in vertical downward direction, due to gravity. Drops occur when packages slip out of workers hands during manual handling, fall from. top of stack during shipping, drop from conveyors due to jamming caused by packages in the front. The second type of impact described as "toss" occurs most commonly during manual sorting operations. The packages are laterally tossed into different sorting bins depending on destination. .As a result the packages remain in a free fall condition, or 16 condition, for a prolonged period of time before impact. The last type of impact referred to as "kicks” occur usually during automatic sorting operations. The packages experience side impacts caused by the swinging arm of sorting equipment or sliding into stationary packages. It is important to understand how these three categories are sensed by the DHR units. For every recorded event, the DHR calculates drop height from both the zero-G channel as well as equivalent drOp height channels. During a free fall, the zero-G channel shows a greater accuracy than the equivalent drop height channels. This is mainly due to the fact that the zero-G channel uses free fall time which varies 25 directly with only the height of free fall while equivalent drOp height channels use velocity change and coefficient of restitution (e) to calculate equivalent drop height. The value of "e" changes with drops on edges, corners, and faces due to the amount of cushioning. Large errors may occur in equivalent drop height calculation even though the velocity change is the same for different events. To categorize ‘various impacts into drops, kicks or tosses, it is important to measure the "Unit Ratio". Unit Ratio may be defined as the ratio between drop height measured using the zero-G channel and calculated equivalent drop height. - h UnitRatio= Zero G Drop Height 8 ,, Equivalent Drop Height 3: . ........ (2-4) Lab simulation of all three categories of impacts was performed and unit ratios calculated for each category. In a free fall drop, the Unit Ratio lies between 0.5 and 2.0 (voss, 1991). In a "toss" the DHR stays weightless for much longer time. The drop height calculated from the zero-G channel becomes very large and inaccurate. However, the equivalent drop height is much lower and as a result Unit Ratio becomes large. Lab simulated tosses showed that Unit Ratios higher 26 than 2.0 are usually common. During "kicks" the DHR calculates equivalent drop height based on velocity change of the impact. However, almost no drop height is measured by the zero-G channel because the unit stays motionless before impact. This results in low Unit Ratios. Lab simulated "kicks" generated Unit Ratios of less than 0.5. Based on the lab simulation tests, the actual data. was analyzed. and categorized based on "Unit Ratios" as follows: If Unit Ratio is < 0.5, the impact is a kick. If Unit Ratio is 0.5 to 2.0, the impact is a drop. If Unit Ratio is > 2.0, the impact is a toss. 2.7 DATA COLLECTION The intent of this study was to measure the shock impact levels in terms of drops, tosses, and kicks, encountered in the overnight parcel shipments across the U.S. Five round- trip destinations were chosen to represent the various geographic regions (Figure 3). The five destinations chosen were: - Monterey, CA - Atlanta, GA — Portland, OR - Rochester, NY - Memphis, TN 27 .0: as s 2265.me £9505 to gas. .0 or as: o cozmczmoo O G i 590 } are \A\ _v. .o___>m_:our (0 $22.35. id: :2 Emcm , z .5628”. mo 55:8 28 .All shipments originated in East Lansing, MI. iMultiple shipments were done for each destination and for each carrier to increase. the reliability of collected data. Five roundtrips were made to each of the five destinations and for each of the two carriers (UPS and Federal Express) for a total of 50 roundtrips. Figure 4 describes the flow diagram of how packages move in the UPS "Next Day Air" and Federal Express "Priority" systems. The Next Day.Air Delivery Service of UPS and Federal Express both use the "Hub and Spoke" system to deliver their packages. Packages were picked up by a courier from the School of Packaging, Michigan State University and loaded in small delivery vehicle referred to as a "Package Car”. The packages were taken to respective Operating centers of UPS and Federal Express in the Lansing area where they were consolidated with all the other packages also meant for next day delivery. The consignment of packages were put into air transport containers which were then transported by truck to the regional air facility. The air transport containers were then loaded into the cargo aircraft which serves as the "Feeder". The aircraft then flew to the national Air-Hub with packages and documents headed for various US locations. These air hubs serve as the central sorting facilities for packages from all over the US. Fig 4: Flow Path of UPS and Federal Express Package Delivery System 30 The UPS air-hub is located in Louisville, KY, whereas the Federal Express air-hub is located in Memphis, TN, where the arriving aircraft are unloaded. The air containers are unloaded and transferred on rollers to the central sort area. Here the employees remove the packages from the containers, scan them, and send them on belts to a central sort area, where sophisticated scanners track and check the package destination and size. Packages speed through the hub on a several miles-long network of belts and chutes. Diverters, activated by information in the bar code labels activate to kick packages down chutes and onto proper sort belts. The packages are then collected by their destination and any special handling that may be required. After sorting, the packages are consolidated together with all the other packages bound for the same destination or service area. These are then loaded into containers and onto another "Feeder” aircraft to be delivered to the destination operating center. The packages, after sorting at the local operating facility, are loaded into "Package Cars" to be delivered to the final destination., The test packages were then return shipped to Lansing, the next day, going through the same process. The entire round-trip duration was approximately four days. The data from.each DHR for each round-trip shipment was uploaded into a personal computer for analysis. 3.0 DATAHAND RESULTS The acquired data from the DHR's was uploaded into the computer and analyzed. The processed data was then imported into a spreadsheet for analysis and tabulation. Dynamic events occurring in the small parcel environment of UPS and Federal Express were separated into drops, kicks, and tosses based on the values of Unit Ratio's. Tables 1, 2, and 3 show dynamic events separated into individual drops, kicks, and tosses for each round-trip to the five destinations. This data has been combined for the two carriers (UPS and Federal Express). The data is summarized into number and type of impacts that occurred during the various shipments and is described in Table 4. There were a total of 2394 dynamic events that were measured in the 50 roundtrips. This averages approximately 48 dynamic events per roundtrip (24 events per shipment). The entire data (Figure 5) consisted of 742 drops (31%), 1045 kicks (44%), and 607 tosses (25%). Table 5 shows the maximum, minimum, and average levels for individual drops, kicks, and tosses measured for each destination. The average drop height for 742 free fall drops was 5.95 inches, the average velocity change for 1045 kicks was 77.91 in./sec, and the average equivalent drop height for 607 tosses was 3.92 inches. 31 TABLE 1. SUMMARY OF ALL DROPS 32 DESTINATION TRIP NO. OF DROP HEIGHT (in.l No. DROPS MAX MIN AVG 8.0. Monterey, CA 1 15 13.90 2.30 6.33 3.44 2 12 20.10 2.20 8.84 5.82 3 13 18.50 1.30 5.85 5.36 4 15 18.70 0.80 5.12 5.03 5 10 11.90 2.10 6.35 3.46 6 16 13.00 1.00 4.96 3.13 7 8 15.10 1.80 6.06 4.90 8 15 25.90 0.70 8.14 7.17 9 28 30.40 0.30 5.09 5.63 10 24 77.80 0.60 10.05 16.27 Subtotal 1 56 77.80 0.30 6.86 7.42 Atlanta. GA 1 4 16.10 1.70 8.90 6.20 2 17 22.20 1.40 6.85 5.48 3 7 13.30 1.50 7.27 3.81 4 14 16.20 1.00 5.31 3.98 5 16 20.50 0.50 5.93 5.52 6 18 1 1.30 0.50 5.36 3.53 7 21 14.90 0.70 4.37 3.13 8 24 22.00 0.80 6.30 4.92 9 9 19.30 1.40 7.83 6.55 10 12 9.70 1.40 5.29 2.84 Subtotal 142 22.00 0.50 5.83 4.19 Rochester. NY 1 12 16.70 0.70 5.00 5.16 2 17 15.30 0.80 4.19 3.38 3 13 13.60 1.40 5.15 3.26 4 14 33.60 0.40 7.69 9.90 5 12 18.10 0.60 5.62 4.65 6 6 18.00 1.20 8.47 6.36 7 13 13.20 0.90 4.62 3.72 8 18 19.40 0.60 5.17 4.37 9 21 13.80 1.30 5.66 3.68 10 22 19.00 1.40 5.38 4.20 Subtotal 148 1 9.40 0.60 5.86 4.47 Note: The average and standard deviation values calculated are for those impact levels that exceeded the minimum threshold level. Most impact data is skewed to very low severity levels and does not represent a normal distribution. This data is, therefore, also presented based on a frequency of occurence versus severity in Table 6. 33 TABLE 1. SUMMARY OF ALL DROPS (Conthuedl DESTINATION TRIP NO. OF DROP HEIGHT Iin.) No. DROPS MAX MIN AVG 8.0. Portland. OR 1 13 18.50 1.50 7.61 4.82 2 14 12.20 1.10 6.07 3.51 3 17 51.40 2.50 10.60 12.37 4 9 5.00 0.40 2.76 1 .41 5 10 7.30 1.20 5.12 2.20 6 22 16.80 0.60 5.65 3.62 7 14 19.30 0.40 4.34 5.61 8 22 20.50 0.60 4.62 4.56 9 34 9.40 0.20 3.47 2.32 10 20 9.60 0.30 3.21 2.03 Subtotal 175 20.50 0.20 4.26 3.63 Memphis. TN 1 5 9.90 2.20 5.10 2.98 2 ‘ 7 15.90 1.00 6.33 5.86 3 6 9.30 2.90 4.13 2.54 4 9 13.00 1.30 6.46 4.55 5 17 17.30 0.90 5.29 4.04 6 20 13.70 0.30 4.78 4.12 7 16 36.50 0.20 7.45 9.46 8 14 18.90 1.40 6.41 4.81 9 15 21.70 2.60 9.04 6.58 10 12 20.50 0.50 7.19 6.46 Subtotal 121 36.50 0.20 6.66 5.72. Total 742 77.80 0.20 5.95 5.13 ORIGIN: EAST LANSING. MI. 34 TABLE 2. SUMMARY OF ALL KICKS DESTINA‘HON TRIP # NO. OF IMPACT LEVEL Iianecl IMPACTS MAX MIN AVG 8.0. Monterey, CA 1 12 139.00 39.00 78.58 31 .15 2 15 1 13.00 35.00 74.33 28.01 3 24 135.00 49.00 86.13 25.36 4 18 155.00 42.00 92.78 37.00 5 17 153.00 39.00 80.76 32.47 6 12 127.00 30.00 66.75 29.29 7 23 180.00 30.00 76.74 36.43 8 25 142.00 43.00 70.16 23.41 9 30 1 14.00 48.00 75.70 18.65 10 16 146.00 32.00 79.63 31.87 Subtotal 1 92 1 80.00 30.00 73.80 27.93 Atlanta. GA 1 7 128.00 42.00 82.71 36.62 2 26 179.00 40.00 80.50 34.71 3 1 1 21 1.00 36.00 106.60 49.29 4 17 180.00 52.00 104.20 38.77 5 25 194.00 44.00 94.68 36.81 6 20 138.00 30.00 82.80 23.36 7 14 206.00 41.00 117.90 42.33 8 19 133.00 42.00 79.05 28.18 9 20 172.00 41.00 83.70 34.67 10 20 177.00 10.00 84.70 39.82 Subtotal 1 79 206.00 10.00 89.63 33.67 Rochester. NY 1 22 149.00 33.00 70.23 29.50 2 19 136.00 18.00 80.90 31.70 3 26 153.00 19.00 68.20 30.60 4 25 109.00 14.00 64.20 23.80 5 25 125.00 24.00 71 .40 26.60 6 16 155.00 33.00 78.81 33.64 7 17 1 12.00 42.00 74.65 18.31 8 25 130.00 30.00 68.16 27.96 9 28 128.00 34.00 67.68 28.42 10 24 141.00 16.00 75.29 30.54 Subtotal 227 155.00 16.00 72.92 27.77 Note: The average and standard deviation values calculated are for those impact levels that exceeded the minimum threshold level. Most impact data is skewed to very low severity levels and does not represent a normal distribution. This data is. therefore, also presented based on a frequency of occurence versus severity in Table 6. 35 TABLE 2. SUMMARY OF ALL KICKS (Continued) DESTINATION TRIP 3 NO. OF IMPACT LEVEL IhJsecI IMPACTS MAX MIN AVG S.D. Portland, OR 1 18 139.00 16.00 79.17 31.61 2 20 122.00 34.00 81.25 24.35 3 21 143.00 17.00 63.95 31.08 4 31 224.00 42.00 82.52 43.20 5 19 133.00 23.00 68.32 29.81 6 29 142.00 44.00 85.76 24.85 7 21 121.00 37.00 71.95 25.52 8 27 126.00 32.00 79.59 27.88 9 22 141.00 36.00 68.09 28.62 10 32 217.00 31.00 76.91 39.01 Subtotal 240 217.00 31 .00 76.46 29.18 Memphis. TN 1 10 138.00 36.00 65.90 28.93 2 12 142.00 12.00 73.75 32.95 3 12 233.00 43.00 108.40 50.50 4 24 153.00 8.00 87.42 38.95 5 22 146.00 30.00 72.95 32.86 6 22 131 .00 28.00 67.68 25.05 7 30 109.00 29.00 67.37 21 .46 8 25 163.00 28.00 71 .76 35.47 9 23 164.00 33.00 85.70 39.44 10 27 184.00 13.00 70.78 34.42 Subtotal 207 1 84.00 8.00 74.81 32.52 Total 1045 233.00 1 2.00 78.33 32.47 ORIGIN: EAST LANSING. MI. 36 TABLE 3. SUMMARY OF ALL TOSSES DESTINATION TRIP # NO. OF . DROP HEIGHT lit.) TOSSES MAX MIN AVG S.D. Monterey, CA 1 5 6.40 1 .60 3.38 1.94 2 12 6.50 1 .50 3.76 1.46 3 11 5.60 1.10 2.75 1.58 4 8 5.70 0.80 2.67 1 .68 5 12 15.20 1.40 4.67 4.76 6 8 15.60 1.70 4.95 4.64 7 14 1 1.00 0.50 4.02 2.79 8 12 11.20 1.20 5.72 3.29 9 16 13.60 0.60 4.32 3.71 10 16 16.80 1.00 3.89 4.16 Subtotal 1 14 16.80 0.50 4.58 3.72 Atlanta. GA 1 3 7.80 0.50 4.10 3.65 2 13 18.50 0.90 3.11 2.54 3 8 31.40 0.70 4.18 3.87 4 16 22.80 0.90 5.24 4.44 5 16 23.00 0.10 4.54 5.37 6 6 7.70 3.00 5.33 2.00 7 20 19.50 0.90 3.79 3.98 8 15 6.50 1.00 3.77 2.00 9 16 18.40 1.40 5.06 4.24 10 14 11.60 1.40 4.99 3.01 Subtotd 1 27 19.50 0.90 4.59 3.05 Rochester. NY 1 10 7.50 1.30 3.12 1.75 2 8 4.50 1.00 2.01 1.07 3 10 6.90 1.00 3.35 2.03 4 13 10.90 1.10 3.97 3.04 5 8 4.90 0.90 3.36 1.38 6 13 21.60 0.60 5.89 6.14 7 15 14.60 0.10 5.41 4.94 8 7 10.70 1.00 4.21 3.13 9 14 15.60 0.60 3.39 3.80 10 15 14.00 0.40 3.39 3.34 Subtotal 1 13 21 .60 0.10 4.46 4.27 Note: The average and standard deviation values calculated are for those impact levels that exceeded the minimum threshold level. Most impact data is skewed to very low severity levels and does not represent a normal distribution. This data is, therefore. also presented based on a frequency of occurence versus severity in Table 6. 37 TABLE 3. SUMMARY OF ALL TOSSES (Cont'nuodl DESTINATION TRIP # NO. OF DROP HEIGHT IhJ TOSSES MAX MIN AVG S.D. Portland. OR 1 9 9.20 0.20 5.40 2.86 2 13 11.90 1.30 4.19 3.13 3 8 10.70 1.70 4.95 3.45 4 9 9.60 1.80 4.77 2.68 5 13 13.60 0.80 4.41 3.70 6 5 7.30 1.70 4.24 2.74 7 20 17.80 0.90 3.07 3.71 8 14 9.20 0.60 3.19 2.90 9 17 10.50 0.40 3.58 2.93 10 14 5.20 0.20 1.86 1.27 Subtotd 122 17.80 0.20 3.19 2.71 Memphis. TN 1 3 7.50 2.40 4.90 2.55 2 9 7.30 0.70 2.90 2.42 3 7 9.30 1.30 4.57 3.47 4 15 17.10 1.20 5.87 5.79 5 20 1 1.60 0.30 3.26 3.20 6 17 5.70 0.60 1.85 1.26 7 17 15.70 0.30 3.52 3.56 8 15 10.80 1.10 3.33 2.75 9 11 3.60 0.80 2.01 0.97 10 17 8.90 1.10 3.90 2.01 Subtotal 131 17.10 0.30 3.39 2.79 Total 607 31 .40 0.20 3.90 2.70 ORIGIN: EAST LANSING. MI. 38 TABLE 4. CUMULATIVE DATA FOR ALL SHIPMENTS I MPACT TYPE "PACT TYPE ROUNDTRIP TRIP TOTAcLYJ KICKS DROPS TOSSES KICKS DROPS TOSSES DESTINATION No. IMPA No. No. No. 96 96 96 MONTEREY. CA 1 32 12 15 5 37.5 46.9 15.6 2 39 15 12 12 38.5 30.8 30.8 3 48 24 13 1 1 50.0 27.1 22.9 4 41 18 15 8 43.9 ' 36.6 19.5 5 39 17 10 12 43.6 25.6 30.8 6 36 12 16 8 33.3 44.4 22.2 7 45 23 8 14 51.1 17.8 31.1 8 52 25 15 12 48.1 28.8 23.1 9 74 30 28 16 40.5 37.8 21.6 10 56 16 24 16 28.6 42.9 28.6 Totd 1 0 462 1 92 1 56 1 14 Avngrb 1 46 1 9 1 6 1 1 41 .6 33.8 24.7 9.0 12 6 6 4 ATLANTA, GA 1 14 7 4 3 50.0 28.6 21.4 2 56 26 17 13 46.4 30.4 23.2 3 26 1 1 7 8 42.3 26.9 30.8 4 47 17 14 16 36.2 29.8 34.0 5 57 25 16 16 43.9 28.1 28.1 6 44 20 1 8 6 45.5 40.9 13.6 7 55 14 21 20 25.5 38.2 36.4 8 58 19 24 15 32.8 41.4 25.9 9 45 20 9 16 44.4 20.0 35.6 10 46 20 12 14 43.5 26.1 30.4 Total 1 0 448 1 79 142 127 Avgl‘l’rb 1 45 18 14 1 3 40.0 31 .7 28.3 8.0 14 6 6 5 ROCHESTER. NY 1 44 22 12 10 50.0 27.3 22.7 2 44 19 17 8 43.2 38.6 18.2 3 49 26 13 10 53.1 26.5 20.4 4 52 25 14 13 48.1 26.9 25.0 5 45 25 12 8 55.6 26.7 17.8 6 35 16 6 13 45.7 17.1 37.1 7 45 17 13 15 37.8 28.9 33.3 8 50 25 18 7 50.0 36.0 14.0 9 63 28 21 14 44.4 33.3 22.2 10 61 24 22 15 39.3 36.1 24.6 Total 1 o 488 227 148 1 1 3 ‘ Avngrb 1 49 23 1 5 1 1 46.5 30.3 23.2 as 8 4 5 3 39 TABLE 4. CUMULATIVE DATA FOR ALL SHIPMEUTS (Conthuedl I MPACT TYPE ”PACT TYPE ROUNDTRIP TRIP TOTAL KICKS DROPS TOSSES KICKS DROPS TOSSES DESTINATION No. IMPACT No. No. No. 96 96 96 PORTLAND. OR 1 40 1 8 13 9 45.0 32.5 22.5 2 47 20 14 13 42.6 29.8 27.7 3 46 21 17 8 45.7 37.0 17.4 4 49 31 9 9 63.3 18.4 18.4 5 42 19 10 13 45.2 23.8 31.0 6 56 29 22 5 51.8 39.3 8.9 7 55 21 14 20 38.2 25.5 36.4 8 63 27 22 14 42.9 34.9 22.2 9 73 22 34 17 30.1 46.6 23.3 10 66 32 20 14 48.5 30.3 21.2 Tetd 1 0 537 240 1 75 1 22 Avgl'l'rb 1 54 24 1 8 1 2 44.7 32.6 22.7 8.0 1 1 5 7 4 MEMPHIS. TN 1 18 10 5 3 55.6 27.8 16.7 2 28 12 7 9 42.9 25.0 32.1 3 25 1 2 6 7 48.0 24.0 28.0 4 48 24 9 15 50.0 18.8 31.3 5 59 22 17 20 37.3 28.8 33.9 6 59 22 20 17 37.3 33.9 28.8 7 63 30 16 17 47.6 25.4 27.0 8 54 25 14 15 46.3 25.9 27.8 9 49 23 15 1 1 . 46.9 30.6 22.4 10 56 27 12 17 48.2 21.4 30.4 Totd 1 0 459 207 121 1 31 Avgl'l’rb 1 46 21 1 2 1 3 45.1 26.4 28.5 8.0 1 6 7 5 5 Grand Totd 50 2394 1 045 742 607 Avg/Tm) 1 48 21 1 5 1 2 43.7 31 .0 25.4 8.0 1 3 6 6 4 ORIGIN: EAST LANSING. MI. Frequency of Impacts (96) 40 50% - Drops Kicks Tosses Impact Type) Fig 5. Distribution of Impacts for All Shipments 41 *Vdu 35.0.? *0. pm swoon—E. =< 0.N F 0. FN 0.m p atteaeeoai 0.500 0.9409 9th LEE—.2 .33. wmmth 9.03. 20.3 mun—>5. .—.U= >¢<5=230 .0" can 6 .m «052“. So: 605.58% 2e 33.? open. on... a 0.0 0.MN p 0.N — coco-3000 $00 0. F _. 0.0.: 0.59 0235000 oamm 0.5 P 0.mhp 0.0N 02.0.5900 $00 0.N 0.50 06 accos— 3 at. a.» 8222 «.0 0.0 «.0 E3555 .v. 5 0.mmu mfih E35532 ...... mummOh .0835 3.0.: ..c: mecca mun—>5. h0¢¢