QUALITY CGNTROL EN m CHERRY PROCESSWG PLANTS Thesis for “19 Degree of M. 5. MICHIGAN STATE UNEVERSITY Donald G. Gillette 11956 .(.'..I WAN 1 <9 £005 QUALITY CONTROL IN SOUR CHERRY PRDCESSING PLANT? Ev Donald G. Gillette A THESIS" Submitted to the College of Agriculture Michigan State University of Agriculture and Applied Sciences in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Agricultural Economics 1956 I. .I52§.";a - ACKNOWLEDGMENTS The author wishes to express his gratitude to all those who helped during the study and in the preparation of the manuscript. Special thanks is given to Dr. Vernon Sorenson of the Agricultural Economics Department for his valuable guidance in the planning, supervising and editing of this study. Many thanks to the Michigan Agricultural Experiment ttation in providing the funds for this study and to Dr. L. L. Roger and the Department of Agricultural Economics for the financial assistance and the valuable experience and training obtained. The help and cooperation of the processing plants and governmental inspection agencies who pnrticipated in this study'hss been a major contribution toward making this study a reality. The author is indebted to his wife, Phyllis, for the encouragement she provided throughout the study and for the painstaking Job of typing the final manuscript. The responsibility for any errors that remein in the final manu- script is assumed by the author. Doneld G. Gillette ii QUALITY CONTROL IN SOUR CHERRY moon-some PLANTS 3! Donald G. Gillette AN ABSTRACT Submitted to the College of Agriculture Michigan State University of Agriculture and Applied Sciences in partiel fulfillment of the requirements for the degree of MKSTDR 0? SCIENCE Department of Agricultural Economics Year 1956 Approved alfaddfl ‘Z X‘MW ABSTRACT The objectives of this study were (1) to analyze and evaluate the present techniques of quality control in sour cherry processing plants, (2) to examine those factors that influence quality, and, (3) to develop application of currently existing statistical techniques of quality control to the cherry processing industry. Thirty-eight sour cherry processing plants in Michigan were sur- veyed during the 1955 cherry pack operations. Questionnaires and plant observations provided information in which to evaluate those factors which might influence quality. Raw product and processed product inspection records were obtained from governmental agencies to analyze the quality received and processed at different plants. The study revealed that there were considerable differences between processors in relation to their attitudes toward quality control and maintenance programs. Many processors attempt to improve the quality that is received by conducting grower meetings, using quality payment plans and employing fieldmen. An attempt to evaluate the effect of these factors on the quality received was unsuccessful. In the processing plants considerable differences exist between the number of sorters placed on inspection belts, receiving, handling, process equipment and ! inspection methods. Sorting labor varied from 1.3 to 6.7 sorters per , thousand pounds.of fruit going over belts in an hour. Time studies also indicate that there is considerable difference in pickout among sorters. Medium sized growers usually delivered higher quality fruit to the processor than did the larger or smaller sized growers. Those growers who did deliver high quality reduced their defects by having less defects over which they had some control. iv The raw product quality received by plants varied considerably from one day to another. Windwhip, scars and undercolor were the more important defects influencing the raw product quality. The grade of the raw product appeared to be a poor index of the finished product grade because of the different quslity characteristics used in grading the two products. The probability of a sample reflecting the actual grade or quality of a lot can be determined by an operating characteristic curve. The increase in reliability that can be obtained by increasing the size of the sample may be revealed with this statistical tool. Quality control charts are decision making tools for management to use in adjusting and allocating resources. ~The control charts described herein are somewhat unique to the general ccncept of control charts. Moving average control limits provide the necessary flexibility needed by food processing industries that must allow for seasonal quality changes for those factors over which the plant has little control. TKBLE OF CONTENTS CHAPTER I INTRODUCTION... ............. ....... ........................ . Quality Standards and Differences..... ....... . ........... . Economics of Quality Control. ........................... .. Objectives of this Study............ ..... .. ......... ...... II, SCOPE AND SOURCE OF mun“ ......... ....... ........ scopeOOCOOOOOIOO....00... OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO O mu Gathered-0.0.0.0.... OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 0 Sampling Procedure....... .................. ..... ......... . Limitations of Study.. .................................... III {HIE CHERRY INDJSTRY. O O O O O O ..... O O O O O O O O OOOOOOOOOOOOOOOOOOOO 0 Importance of the Industry...... ..... ............ ........ . The Processing Industry..... ............................. . Cherry Products Produced......................... ....... .. mrket outlets.’ ..... ooooooooooeooooogeso ........ O ...... 0. Inspection and Quality Standards... ........... . ........ .... IV FACTORS INFLUENCING RAW PROHJCT QUALITY......... ........ ENEMLANALYSIS.OOOOOOOOOOOOOOOOOOOOOOOO.0000.0.0.0. 0000000 Transportation and Handling Methods....... ....... ......... MethOd Of Insoec tion. 0 O 0 O I O O O O O O G 0 O O O O I O O I C 0 O O O O O O O O O O O a 0 O Locational Influence..... .............. . ................. . Pamfint Programsooeeo ooooo so eeeeeee eeeA ..... o eeeeeeeeee sees Educational Programs........ ............... ...... ....... . Importance of Fnctors in Determining the Grade of the M“ PrOduCtOOO OOOOOOOOOOOOO 0.0 ....... OOOOOOOOOOOOIOOOOOO mREE PMNT AMLYSISOOOOOO ...... 0.. ..... O O 0000000000000000 0 Seasonal and Daily Quality Variations....... .............. Quality Differences Among Growers” ..... ............ ...... Size or GrowerOOOOOOOOOOO ..... .0... ..... 0.0.0.... ........ 0 Importance of Specific Defects in Determining Raw PrOd-uct WlityOOOOOOOOOCOOOOO. ........ OOOOOOOOOOOOOOOOO V INPLAN‘I' QUALITY WAC'IOR ANDOUALITY RELATIONSHIPS: ..... . ..... INPLANT QJALITY FAcmRS‘OOOOOOOOOOCOOOOIOOOOOOOOOOOOOOOOOOOOO Handling Methods in Receiving........... ....... .... ...... . sorting PrOCOSSOOOOOOIOOOO ....... .00. ........... 0. 0000000 O Fitting Operation........... ............ ..... ......... .... Hotpack Process.............. .................. . .......... 601(3ka Processonooeoeeeeeeoeeeosooeeoeeoeooe eeeeeee eeeol InspGCtio‘ MethodSQOOOOOOOOOOOOOOQOO...O.0..OOOOOOOOOOOOOO QUALIN RMTIONSHIRO I O O O O O C O O O O O O O 0 O O ....... O O O O O O O I O O O 0 O 0 Raw and Final Product Quality............................. vi F'P'h‘ P‘ haha \o-qxn \n tptp\0\o \o ~qrpra MN MO 28 TABLE OF CONTENTS - Continued CHAPTER Page Influence of Raw Product Defects on Processed Pr0duct Qm‘lity................... ...... 0...... ........ O 66 Importance of Factors Determining Final Product §COre..... 67 Daily Changes in Finished Product Quality...... ........... 7O Drained Weights... ....... ... ...... . ....................... 70 Probable and Final Scores.. ............................... 71 VT SAMPLING AND QTATIgTICAL QUALITY CONTROL......... ........ ... 73 Evaluation of Raw Product Sampling......... .............. . 73 Evaluation of Processed Product Sampling .............. .... 78 Theory and Application of Control Charts...... ............ 82 VII SUMMARY.AND CONCLUSIONS. ............................. . ...... 100 BIBLIOGRAPHY.. ................. .. ....... ......... .................. 105 APPENDICE 0000000000000 O 000000000000 O O O O 0 C O O 000000 O 0000000000000 O 0 O 107 vii TABLE II III IV VF! VFB VI VII VIII IX XI XII XIII XIV LIST OF TKBLEU Analysis of Cherry Pack Sales - 27 Plants (195b)............. Score Chart for Canned or Frozen Sour Pitted Cherries........ Average Raw Product Quality (percent U.S.. No. 1) Received by 27 Processing Plants in Michigan in 1995 - by area... ........ Payment Plans for Raw Product Cherries....................... Grower Differences - Percent Controllable and Noncontrollable Defects - Plant I....... C ......OOOIOOOOCOCCOO0.......0...0.... Grower Differences - Percent Controllable and Noncontrollable DfifeCtS - Plant IIeeeeeseoooeo Grower Differences - Percent Controllable and Noncontrollable DefeCtS - Plant IIIoeooooe ' Relative Importance of Specific Defects - 3 Plants... ........ Sorting Performance - 17 Processing Plants................... Cost of Sorting Labor per Thousand Pounds at Different Volume Levels and Different Number of Sorters (wage rate $1.00/hour) Cost of Sorting Labor per Thousand Pounds at Different Volume Levels and Different Number of Sorters (wage rate .85/hour).. Man Hours of Sorting Labor Required to Up Grade Fruit Using Average Pickout Rate............. .........0.0000000000000000. Importance of Raw Product Defects in Determining Processed Product Quality at 17 Plants......... Relative Importance of Color, Defects and Character in Determining Finished Product Quality............ ......... .... Changes in Scores — Probable to Final.................. ..... . Scoring System for Pinished Product... ..................... .. viii 33 an #0 #1 #2 52 5a 55 56 66 68 72 80 FIGURE NUMBER 1. LIST OF ILLUSTiATIONS Location of Cherry Processing Plants Included in Survey..... Red Cherry Production in Michigan (1990-55)................. Red Cherries - Michigan's Share of U.S. Production 1918-1955 Daily Raw Product Quality variations - 3 P1ants............. Average Percent Defects - By Grower Size - By Plant........r Influence of Raw Product Quality on Finished Product Score . Plant100.000.0000....000.00000000000000000000 ...... 00000000 Influence of Raw Product Quality on Finished Product Score Plant I100...00000000900000.00000000000000000000000000000000 Influence of Raw Product Quality on Finished Product Score Plant IIIOO0000000000000000 00000 000.0.0000000000000000000000 Accuracy of Different Sampling Plans in Determining Raw Preduct Acceptability000000000000000000000000000.00I00000000 Confidence Ranges for Raw Product Sampling.................. Grading Probabilities on Finished Product................... Control Chart Based on 3 0".............................. Control Chart - Total Scores - With Trend................... Control Chart - Total Scores - Moving Average............... Control Chart - Color - Moving Average...................... Control Chart - Character - Moving Average.... ......... ..... Control Chart - Defects - Moving Average.................... Control Chart - Net weight (103 cans)......... ........... ... Control Chart - Drained Weight (103 cans)................... Control Chart - Vacuum (303 cans)........................... Operating Characteristic Curve for a Control Chart (with Upper and Lower Control Limits)..................... ..... ... ix Page 10 16 16 38 h# I 64 64 65 75 77 81 an 88 91 91 91 91 94 . 95 95 98 CHAPTER I INTRODUCTION Quality §tandards and Differences The emphasis placed on quality control and improvement varies widely among plants processing and marketing Michigan red sour cherries. Many packers concentrate on quantity production at the lowest possible cost. This often means packing a lower quality and selling at a lower price. Other packers attempt to build a high quality reputation for their business in an effort to increase sales and maintain steady outlets for their product. Improving the quality of the finished product starts before the raw product reaches the plant. Grower education through fieldmen, grower meetings. mail contacts and other services are used to help the grower produce better quality than he would otherwise produce. Payment plans that offer premiums or discounts are used to provide the grower with an incentive to improve quality. New methods of transporting the cherries from the orchard or receiving stations to the plants are being attempted by many plants or growers in order to maintain incoming quality. From the time sour cherries are received at the plant. many steps occur in which the quality of the product can.be either improved or lowered. Improper receiving and handling methods may cause bruising. extremely long soaking periods may cause water scald, and improper mechanical equipment may cause unnecessary quality deterioration. Pbor working conditions. too few sorters on the belts, lack of eupervision and poor inspection procedures all may result in inadequate quality maintenance. Wide differences exist between plants in relation to the means by which they attempt to improve and maintain quality. Many of the larger plants hire specialised personnel for quality control work. Smaller plants are usually much more limited in quality control pro—~ grams because of the lack of specialised personnel. Extreme differences in methods of quality maintenance and control exist in plants of equal eise. Lack of management ingenuity and unwillingness or inability to acquire capital or belief that their plant is already operating very economically and efficiently may be a few of many factors that delay possible plant improvements. Some managers appear to be aggressive in developing and using new tech- niques to improve quality. Other managers are slow in adjusting to the changing times and the new technologies that become available. new equipment that will improve quality control or maintenance is often disregarded by the processor because of price uncertainty for his prod- ucts or because a shortage of capital prevents making the investment required. Other managers are unwilling to admit that their plant operations can be improved. Iconqmicl of Quality Control Business firms often can increase sales without lowering price. either by altering the buyers attitude toward the product. or by modifying the product so that it conforms more closely to what the buyer wants. Sales promotion and advertising are the common means for changing the buyers.attitude toward a product. Product modification generally involves making changes in some observable characteristic of either the product or the container. or a change in the service asso- ciated with the product. Sales based on maintained quality calls for brands as a means of identification to guide the consumers. In the processing industry quality-brand buyers will be attracted to those plants which have gained a reputation for packing a quality product. The word quality can'be defined in very broad terms or in more specific terms. Abbott defines quality as “any or all of the various qualitative characteristics of a physical product or service or com- bination of the two offered for sale.'1 In this context quality means any characteristic of a product that might-affect the price that a buyer is willing to pay or the quantity he will take at a given price. Frank Knight writes: 'If people are willing to pay for 'Sunny Jim! poetry and 'It floats' when they buy cereal and soap, then these wares are economic goods. If a name on a fountain pen or safety razor enables it to sell at a fifty percent higher price than the same article would ordinarily fetch. then the name represents one third of the economic utility in the article, and is economically no different from its color or design or the quality of the point or the cutting edge or any other quality which makes it useful or appealing."2 Quality as used in this thesis is restricted to those character- istics of the raw and processed contents in the container that influence the demand curve facing the firm. Quality will refer to the color and appearance. size and shape, uniformity and flaws in the raw and the recessed cherries. n-ained weight and fill weights are also 1L. Abbott. Quality and Competition. Columbia University Press, New _Ionk, 1955. p. U. Zr. E. Knight. Risk ,Uncertailgz and Profit, Boughton Mifflin Company, New York, 1921. p. 262. to be considered as quality factors. Many producers desire to improve their positions in the markets and do so by taking independent action to modify the quality of their respective products. Quality may be improved or lowered, depending upon the kind and amount of resources used in processing and handling. The grower may employ pickers who 'stripP trees rather than picking the cherries selectively; One processor may have only two sorters for every thousand pounds of cherries passing over the inspection belts in an hour while another processor receiving identical quality fruit may have six sorters per thousand pounds. Quality improvement usually involves a greater utilisation of re- sources and this usually means a greater cost. The additional resources used in improving quality must be paid for and therefore the processor believes he should have a higher price for his product. Several packers stated that they were not attempting to pack A grade because the price differential between A and c grades is not ordinarily great enough to warrant these additional costs. Quality competition may prove to be more effective than price competition in increasing the sales of a firm. Ability to sell the cherry pack the same year it is produced is often very important to small processors. The need for money to meet current operating expenses may involve costly loans and storage unless the processed product is sold immediately; By maintaining quality product standards a processor helps assure an outlet for his product even in years when part of the total pack is carried over. The benefits of quality control are often hard to measure in monetary terms. Decisions involving quality are often made on an intuitive basis without attempting to measure the result in dollars and cents. In making economic studies to guide decisions as related to quality, Grant divides the influencing costs into production, acceptance, and unsatisfactory product costs.3 Production costs refer to those costs involved in the production of the product under consideration. The different amounts of labor requirements etc.. to bring the raw product up to a given quality. Acceptance costs refer to the testing and inspection costs and the costs for administering the acceptance program. Raw product and finished product inspection costs are included in this classification. Unsatisfactory product costs refers to the costs incurred for accepting a product that turns out to be unsatisfactory for the intended purpose. These costs can.be incurred by obtaining raw product quality that is below acceptable quality standards and by loesing sales because of selling products that are not of acceptable quality to the purchaser. Many of the costs of quality control are incurred either to correct mistakes or to police them. High costs of increased inspection, processing losses, and unsatisfactory product may be substantially reduced by an effective program of statistical quality control. In the cherry processing industry, for example, plant or government regulations may require that all or some specified percentage of the processed product contain a stipulated weight. Most of the processors inter- viewed indicated that they overfilled containers to avoid trouble with regulatory agencies and to guard against unsatisfactory buyer acceptances. .- ‘q vw—‘i—v vi :3]. L. Grant, Statistical_guality Control, McCraw-Hill Book Company 'Inc., New York, 1952, pp. th-th. The resulting average overfill may often prove to be an expensive insurance cost to the processor. For an example, assume that a plant is packing #10 cans and the buyer specifies 108 net ounces per can and the packer attempts to maintain a 2% overfill. _Quality control tech- niques may show that little variation exists in can fill and the 2% overfill is excessive. At 10d per pound for fresh cherries, the reduction of 21 on the fill weight of a thousand cases of #10 cans will result in a savings of about $60.00.“ Quality control techniques and procedures may be extremely useful in making decisions on the quality levels to be maintained. The primary purpose of the control chart is to improve decision making. Control charts not only show where quality improvements can be made, but also show where reduction in quality levels may be desirable. If quality is extremely high it may not pay to allocate as large a quantity of resources to quality maintenance. If the quality level is extremely low, more resources may be needed to improve quality. Statistical 'quality control is a tool to aid the packer in making these economic decisions. Quality control and maintenance is a subject of increasing impor- tance in the food processing industry. State and regional conferences and clinics and trade Journals are emphasising quality as a means of increasing sales.5 1"Assuming that 120 cherries . 1 lb and a #10 can weighing 108 02. con- tains 700 cherries by count. 5V..A. Gould, Quality Emphasized at State Meeting, Food Zackers, March 1955, pp. hh-h6. In Michigan the cherry processing industry has become increasingly concerned with quality maintenance and improvement. Increased cherry production has brought the need for increased sales and many packers are looking toward quality improvement and the development of new cherry products to absorb the expected additional output. Ob actives of this Study Because of the importance of the cherry processing industry in Michigan and the interest in quality control, the objectives of this thesis are as follows: 1. Botswana 9n pf thslqhassslulquaet. nvlmgininslm pgggggsigg operation: Raw product quality received by plants will be compared with the quality of the resulting processed product. 2. Determinatigufihe accuracy of current qualityMurement techniques: Analysis of the current inspection procedures used by different plants and development of operating characteristic curves for different sampling plans. 3. Eraluagion of inzplant operagigg practices whigh_gffggt_pgoduct qualigz: Receiving and handling methods, inspection and sorting rates, machinery and equipment, qualityflhuantity flow adjustments, and actions of supervisory personnel, etc. 1+. Matias of factors influencipg_t_he quality 0&1th received: Inspection procedures, locational influences, payment programs, educational programs and size of grower. 5- Mmluafisuuetvwn simulzeniniimz 31" of grower, seasonal and daily quality variations, importance of specific defects, and transportation and handling methods. 6. Application of control chartg_tg;qgalitz maintenance; Evalup ation of the quality control technique. Application of this technique to a typical cherry product. usefulness of this tool in the processing operation. CHAPTER II SCOPE AND SOURCE OF MTA Scopg Data for this study were obtained through a survey of 38 sour cherry processing plants in Michigan. Most of these plants were visited twice: first for an interview with the plant manager, and second, for a visual inspection of the plant while processing the cherry pack. The reason for two visits were: (1) for a more complete interview with the manager before he became too involved in the processing operation, and (2) to hasten the visits when the processing season once got under— way. Regardless of this precaution, several plants were not seen in operation because of the short processing season that occurred in many areas in 1955. The processing plants included in this survey are located in twelve western Michigan counties (Figure 1). Of the 38 plants, 26 were inspected while in operation. The output from these plants represented over 95% of the total tonnage of sour cherries canned or frozen in Michigan. Data Gathei:g Different types of questionnaires were used depending on whether the plant used government inspection services or maintained its own inspection service. The questionnaire for plants withoum continuous inspection contained more detailed questions referring to inspection procedures. The survey included eleven plants with continuous-in plant ., ' ”mum.“”,,_,,,_,,.n.,.muu...m-ullhl'llmummm n 0".-0 -mQI'U sense 10 Location of Cherry Processing Plants Included in Survey FIGURE I /W /" l m [5 {£53 1.26%: L! -..] . ! iMm i fi;:m"[‘? L L" 'L' -- --l w/M 111:? ., ""‘ ”a" Clem? !OHPPEWAM ' \. I L, i i ~ "-~ \.‘\b\f\‘ !L cxmsou‘: mi£fi__~f ! iMACKINuCQu: ...... ...... ”4‘ i“ TJ ‘ ...-d, f \ “MT-t J Wham? 17!] 7 I J (74" 90% 8 (En:\\% Maori?" two. 55:?) »mr"gc:;7- -TALPENAV\1>& :5 “WIN ‘ : i f ? mm Lin—aim» (ream mm ' I o magi 4mm om. aw ”‘IOSCO l ...1" MICHiGAN o... . , ..L ..l . (Mason MK! TOSCIOM F413}. .cuowhW' '30 ! i I F}, rim 0 ' DAY )girm mtfismrimm frame- I I I £552.. Wei: 0 ‘0 . . . inEfia‘r !GQA 1710;! lufiifi. ! l \ ' i i '5: .__ T [Juan ! ' “N “5 . m - 0 [75m rein FGV—i’mnmfiic F: a“ . C O ! i ! s ..L .,.L.. ..1.. l -J... ....L- 5mm [PM] awn Wm I. mum kimono}?! ' s ‘ I ! t e c—L—~. ...—...... o - -..—u-‘L. Ismail. amour firicusaw ' 4%nmw [mm . g 1 !! I 0 , .... -._._.~L._...- L- ...-L, E A hm Masons 'lflMWL'f-‘j W E O O ; HI‘IIOOHU 0" '5. o 4‘ b ' . '3 I t . Os l-o-.-e—“ ' 11 government inspection and twenty-seven plants without this service. The purpose of the visual inspection was to observe the processing operations of each plant in order to make between plant comparisons. Time studies were made on the rate of pick out from the sorting belts and the volume passing over these belts. Information was gathered as to the types of equipment used and the labor used in maintaining quality. Raw product inspection records were obtained on all sour cherries that went into processing plants in Michigan during the 1955 cherry season. Michigan law requires that all cherries to be processed must be inspected by the federal-state inspection service. The records obtained from the inspection service included: (1) name of the grower, (2) name of the inspector, (3) processing plant, (a) location of inspection, (5) date of delivery, (6) size of load, (7) site of sample taken, (8) specific defects, and (9) the grade and score of the lot. Permission was obtained from several plants, which maintained government inspection service to obtain data from finished product inspection records which are on file in the Detroit office of the Processed Fruit and Vegetable Idvision of the usna. Information con- tained on these records included grades on lots and individual samples, net and drained weights, vacuum and head space, the name of the inspector and the date and time the samples were drawn. The points system as used in determining grades of the processed product was also contained on these records. The importance of’colo; character and defects in determining the processed prdduct grade was noted by the relative weighing of these points. 12 Samplingrgrocedure Raw product records were analyzed by taking 100 samples from all inspectors at each inspection point. These samples were used to deter- mine the average quality raw product going into each of the plants and regional and local differences. All of the raw product records from three plants were analyzed to determine the extent of variation between growers and to provide data for determining the correlation between the quality of raw product received and the finished product which was produced. These records also provided information on the day to day variations in quality of the raw product received. Final inspection records were obtained on three plants. All of the records available from these three plants for the 1955 pack were used to determine quality improvement. Probable grades were also obtained on most of the processed product from the three plants. Probable grades are taken immediately after packing while the final grade is taken after the product has been in storage for a period of time. Limitations of Study The material contained in this thesis provides the researcher with °a good background for further study of quality maintenance in the sour cherry processing industry. General comparisons are made of the individ- ual plants in the cherry processing industry. More detailed comparisons were made of three of these plants. The information contained in this paper does not provide a detailed comparison study of all factors that would affect quality maintenance. More detailed study of in-plant and producer oriented quality maintenance is needed of plants in the industry. 13 Much of the information obtained in the questionnaire and in the plant observation is subjective. It is highly probable that many managers over-emphasized their quality maintenance operations while other managers may have underestimated theirs. 'Opinions formed by the author were affected by not only the interview with the manager but also by the observation of the plant operations. Most plants were observed for about two hours. This provided information only as to major differences in operational methods that might affect quality. The time studies taken of sorters on the sorting belt to determine the rate of pick out was affected by the volume and quality passing over the sorting tables at the time the observations were made. The way the plant was operating during the short period of observation in some cases may not represent an accurate picture of its usual operations. A more detailed study of these plants would provide this information. Quality improvement as measured by the change from the raw to the finished product is difficult to measure because of the dissimilarity of the factors determining the raw product score and the factors determining the processed product score. The probability of error that 'is inherent in sampling provides the basis for further questioning as to the reliability of this measurement of quality improvement. Records were not obtainable on all of the finished product inspections for those plants in.which quality improvement was measured. This often resulted in only a small proportion of the finished product being compared to the raw product scores. Quality control techniques that are presented in this thesis can be used by all processing plants. Sampling is as much a part of this technique as the actual control technique. These tools tell manAgsment 1“ when adjustments are needed to improve or maintain quality. Quality control technique is merely a guide to be used in operations. The adjustment must be made by management. “4&5.- CHKPTER III THE CHERRY INDUgTRY The purpose of this chapter is to develop an understanding of the industry in which the quality problem will be discussed. A knowledge of the characteristics of the processing industry is essential to the understanding of the quality problem. The types of cherry products produced and their outlets are additional background information that ”ii__4‘ will prove to be useful. ‘ . 5 -r. Inspection and Quality Standards are contained in the latter part of this chapter. This material is necessary in developing the quality control problem in later chapters. The attitudes and opinions of the processors toward the present inspection and quality standards of both raw and processed cherries are discussed briefly. Imtancs of the Industgz From 1950 to 1955, 603% of the total red cherry crop of the United States has been produced in Michigan. The Michigan Department of Agri- culture predicts that from 1955 to 1961 this state will increase the number of red cherry trees planted by 28%.1 Based on normal expected yields per tree. the state crop is expected to total approximately; 100,000 tons by 1961. Any substantial change in production technology can by expected to boost this production still higher. The red cherry industry has become localized and specialized A A“ 1This estimate was based on a cherry tree planting survey made by the Hichigan Federal State Crop Reporting gervice in 1955. 15 16 Red Cherry Production in Michigan 1930 - 1955 FIGURE 2 I" 1 t ’ t I 100. 90. 80. 70m econ. no vagueness ] Wtj. Ed I 6 3 1935 1916 1933 1395 ‘19uo Ye. a rs 1930 'Red Cherries - Michigan's Share of U.S. Production 1938 - 1955 FIGURE 3 ..W n fl .— q 7 q QM. qq 0v 6 5 3 2 1 coaguaeonm .w.= Heaps co m mama emafl mmmfl mama Hmma onma mama mama mama mama mama esma mama «:mH Hama oamfl ommfi mmma )Qsarx 17 through time. In Michigan the major producing areas are concentrated along the shores of Lake Michigan. The economic activity of many communities are heavily dependent upon this industry. The 1949 cherry survey estimated that there are aboum b.000 com» mercial cherry growers in Michigan.2 The large commercial cherry growers are concentrated in the northwest and central west districts in the state. The average orchard size in the northwest district is 1,150 trees as compared with an average of 350 trees per orchard innthc south- west district of the state. The southwest district has much more diversified fruit and vegetable farming than the central and northwest districts and is therefore, not as dependent upon the cherry market. The Processing Industgy In 1955 there were #1 firms or an plants engaged in the canning and freezing of red sour cherries in Michigan. Seven of these firms processed over half of the quantity produced in the state. The smaller plants are concentrated in the southern part of the state.. Twelve plants are located in the northern district, twelve in the central dis- trict and twenty in the southern district. (See figure 1.) Those plants surveyed varied in the number of years that they had been processing from 2 to 78 years. Of the thirty-eight firms sur- veyed. thirty-two are corporations, three are cooperatives. two are single owners and one is a partnership. Management in the processing industry appeared to have little turnover. Twenty-two of the managers contacted have served in this capacity since the origination of their '— — 2Michigan Federal-State Crop Reporting Survey, 1949. 18 processing plants. filong‘with the increase in production of red sour cherries there has been a small amount of expansion in the size of processing plants. Boger points out that there has been a tendency for the range in the relative size of individual firms to become smaller.3 Because of erratic crop conditions due to frost. hail, windwhip, etc.. many processors have not expanded their plant sise. One year a processor may lack adequate facilities to handle the crop and is forced to operate Zh—hours a day. The next year this same processor may need only a skeleton crew to handle the cherries that are brought to his plant from the surrounding area. In 1955. two plants that ordinarily process red cherries did not operate because of the poor cherry crop in their area. Other plants reduced their red cherry processing tine from the normal four week 'period to less than two weeks because of the poor quality that was to ‘be had. Over 90 per cent of the sour cherries produced in Michigan are processed. This emphasises the importance of the cherry processing industry as a market for the growers product. To help insure an adequate supply some plants in the state receive and deliver the raw cherries to their plant from receiving stations over 100 miles away. In years of extremely good yields these processing plants help to relieve the surplus problem in areas where the local processing plants cannot take care of the crop. During 1955. some plants that were in the midst of a large crop area contracted with other plants to take some of 3L. L. Bogsr.|!ighig§n Red Cherry Pricgg, Special Bulletin 371, Agri-’ cultural Experiment Station, Michigan State University, East Lansing, Mdchigan, 1951, p. 1b. 19 their growers cherries. The southern plants generally haul from distant receiving stations or plants because of the desire to extend their processing season once they have the labor available and the plant organized for cherry processing. The desire to extend the processing season may also depend on.what other crops the particular plant may plan to process. In most k cases it can be said that the red sour cherry.is the main processing T; crop for the plant though it is much more important to northern plants than to southern plants. Cherry Products Produced uJ The cherry processing plants in Michigan can be classified according to the kind of pack produced. 0f the 18 plants surveyed twenty plants 'produced both hot and cold packs, eleven produced hot pack only and seven produced cold pack only.“ Many of the newer plants in the industry tend to specialize in coldpack Operations. In 1935 the percentage of the total pack frozen in Michigan was less than 10 percent. In 1955 10 percent of the total cherry pack was frozen. Even though the expecta~ tions of some individuals have not been realized, the coldpack has increased in importance as a method of marketing and processing the cherry crop. Canned cherries are generally packed in No. 10, No. 2, and No. 303 containers. The No. 10 can is a larger container that is used in the institutionaltrade. The no. 2 and No. 303 cans are consumer sized for __A_~. “Hetpacking is commonly referred to as canning. It is a heat treatment process in which the contents in a container are cooked. Coldpacking refers to the process of freezing the contents in the container. 20 distribution in the retail channels. Michigan processors have canned about 69 per cent of the actual cases of red sour cherries since 1945? Coldpack cherries have been directed primarily toward the institutional trade. The large 30 1b. tins is the most common size used. Fifteen pound tins and two pound cans are also used as containers for cold packing by some plants. The lack of cold storage facilities and the cost of acquiring these facilities has definitely been an inhibiting factor on the expansion of the cold pack market. New red cherry products are being introduced by the industry in order to further stimulate demand. Syrup packs are being put out by several plants in both hot and cold packs. In addition to taking advantage of lower drained weight requirements for syrup packs, the processors hope that the consumers will find the pack more useful for 'purposes other than baked pies. Cherry products are being developed and promoted by the Rational Red Cherry Institute for desserts, toppings) Jams, Jellies and sauces. With the expanding production, there comes the need for expanding demand. Market Outlets Processing plants sell their finished product either through brokers or direct to buyers. In analyzing the sales of twenty~seven plants, it was found that two-thirds of the sales are made through brokers and one-third of the sales are made direct. A greater percentage of the cold pack sales were made direct than.were the hot pack sales. 5 . WM. National Canner's Association, Washington, D. 0., June 1956. 21 No consistancy was found in the type of sales made by the various plants. Some plants sold their entire pack direct while others sold their entire pack through brokers. Cherry buyers in order of importance are wholesale houses, food processors, chain stores, institutions, and the government. Regardless of whether these sales are made direct or through‘brokers their relative importance as sale outlets remained the same except for the government and institutional sales. TABLE I Analysis of Cherry Pack Sales6 (l95b) 2? Plants —.—~.—_-.-—-— - ...... Hot 4:. Cold Pack «g of Sales tannin ‘i‘fhole soon. %Inst. $Food __firsale_ fig _§roc. Direct 32 20 #2 6 5 2? Broker ___ 68 __ 1.5 1+1 2 _ 7 “fi 33 n1 Sales 100 16.___152.7 3.3 6. 3 11.1 Cold Pack "‘ "'"" w'" Direct #0 5 31.0 7 7 Broker 60 _ __ 1 _Q,O 2 8.8 9.6 Cold Pack Sales 100 2.6 13.8 1.2 8.8 62.6 Direct 29 34 1.7 _ 11 2 6 Broker ____ 71 ..._.__ 24 67 _.‘ 9 :3-._ Hot Pack 3&1» 100 26.9 61.2 3.2 “.3 3-0 H # m Cold pack sales are mainly to food processors while most hot pack sales are to the retail trade either through wholesale houses or chain stores. A considerable volume of the cold pack tins are sold directly to consumers by many of the smaller processing plants. The smaller sized containers of either the hot pack or the cold pack are sold to wholesale houses or chain stores. The larger sized containers are sold to govern- ment agencies, institutions or food processors. ....“ 6Th690 percentages rebresent‘neerage percents of the sales of 27 plants and not the percent of total pack sales for the packs. The average sour cherry processing plant sells 8h percent of its cherry products to the same buyer year after year. At the extremes one plant sold only 50% of its products to the same buyers and two plants sold their entire output to the same buyer year after year. No relation- ship was apparent either between the percentage of steady customers or the type of outlets for an individual processor and the quality maintenance ET} I program used. Inspection and Quality Standards Raw product cherries that are to be processed must be inspected by J‘ f‘.‘ the Federal-State Inspection Service.7 Under a cooperative arrangement the inspectors are hired by the state and work under the supervision of the United §tates Department of Agriculture. The inspector's report shows the percentage U.S. No. 1 grade cherries, percentage No. 1 Michigan Revised grade standards and the specific defects found in the sample.8 Samples are supposed to be drawn randomly so that the quality of the sample will be as closely representative of the quality of the growers load as possible. .Usually a 500 gram or 1 lb. sample is drawn and the score is determined by weighing the defects. The Michigan revised standard is used to determine the acceptability of a given lot for processing. In Michigan no plant is allowed to process “-..—.... .— -...- ——.—.-~-. 7In 1955 the Federal-State Inspection Service maintained 108 inspection points and 100 inspectors in Michigan for the inspection of sour cherries. 8Standards for Red Sourmgherries for Manufacturegs. U.S.D.A. P.M.A. (Effective April 20, 1941.) Michigan revised standards are identical to U.S. Standards except for the exclusion of the specific defects of lugscald, stems, undersise and under color. 23 somr cherries that grade below 88 score.9 Most cherries that grade below this score are used for Juice. The U. 3. Standards are used by many processors as a basis of payment for quality. Many plant managers reported that they were dissatisfied with the sampling done on the growers loads to determine its grade. Some felt that it is economically impossible to arrive at an accurate grade on these loads. Other packers voiced much concern over the growers practice of taking cherries that were turned down for processing at one plant to another plant where they were accepted. . About half of the processors said that the Michigan revised standards are too low. The processors stated that the omission of color, under- sise, lugscald and stems from this standard encouraged poor harvesting and handling practices on the part of the grower. Processors argued that the standards used on finished products did not exclude those factors which are excluded in the raw product, and thus, the special concessions given to the raw product standards are not Justified. The registration of complaints by plant managers against the omission of quality factors in the revised standards are as follows: Color 21, size 12, lugscald 1h, stems 1h, sugar content 2, and other 2. Undercolor and sugar content are in most cases the results of immature fruit. Many growers picked their fruit as early as possible to reduce the danger of windstorms or hail damage and as a result packing began with under ripe fruit. This resulted in lower drained weights, smaller cherries and poorer color in the processed product. 9Special permission was given to process sour cherries in 1955 that graded below 88 providing they brought this grade up to 88 on the sorting belts. i-I'Wfl~ «an: " .l r’ The finished product is inspected either by government inspectors or plant inspectors. The U.S.DMA. furnishes the food processing in- dustry with inspectors who are allowed to certify the quality of the 10 product according to U.S. Standards. This service is open to all plants who apply and meet the requirements that are set.11 Plant inspectors do not certify the quality of the finished product but in ?“‘ most cases furnish the plant with reliable records as to the quality of certain lots. The U.S.D.A. provides inspection procedures which inspectors must follow in performing their duties. Many plant inspectors remove samples Laf from the processing line at certain time intervals rather than basing their sampling on quantity rates. The minimum sampling rates of the Agricultural Marketing Service, however, provides a fairly close approximation of the sampling done by most plants.12 In the grading of the finished product on cherries the scoring system shown on the following page is used. This system helps in determingflthe grade by giving respective ratings of the factors color, defects and character in addition to the other requirements (such as size) as defined in the grades. These factors and their relative importance in determining the grade expressed numerically in the chart on the following page. M (on Section-52.2112). (era Section 52. 21a. 110. S. D>.A. P. M. A. Questions and answers on Government Inspection of Inspection of ProcessédJTTuits and Vegetables. 12Agricultural Marketing Service Standards Inspection Marketigg . Deps.rtment of Agriculture Public Law 156, 83rd Congress. Approximately July 28, 1951, p. 5. 25 TKBLE 11 Score Chart for Canned or Frozen SOur Pitted Cherries Factors Max. Pts. Gd. A Fanqy Gd. C. Stand. Gd. D substand. Color 20 17-20 14—16 0~13 Defects b0 3h—40 28-33 0-27 Character #0 3h-40 28-33 0-27 Minimum Score 85 70 The total number of points that can be scored for any lot is 100. If any of the three factors fall into a lower grade based on the above points system then the grade for that nroduct Will be the lower grade regardless of the total score. In addition to grade specifications, the standards for the canned Droduct contain recommended brix measurements for syrup packs and drained weights. Neither one of these factors are incorporated in the grades of the finished nroduct and are not considered as being factors of quality for the nnrnose of these grades. Most pickers, however, follow the recommendations for drained weight, fill weight and syrup content provided in the Canning Trade Almanac}.3 All sour cherry processing plants use the U.S. Standards for inspecting and grading their finished products. These standards pro- vide the cherry packing industry with e.yardstick with which thqy can compare and measure the quality of their product. Grade certification of the finished product by governmental inspectors has been of great assistance in the financing, storing and marketing of the cherry peck. 13Canning Trade Almanac is compiled and published each year by the Canning Trade, a weekly business Journal of the canning and allied industries. .....-l.’ b. in! 26 Wholesale, institutional, chainstores and other buyers frequently demand products that have been graded according to the U. S. Standards to enable them to purchase the quality of pack desired. In Michigan, eleven sour cherry processing firms acquired govern- ment inspection service for continuous or in-plant inspection in 1955.1“ Fifty per cent of the firms which did not have continuous or in-plant inspection indicated a desire to have such service either now or in the near future. These firms gave the following reasons for not having this service at the present time: . Cost F.” Annu- hL. a l 2. Size of Operation 3. Viewpoint Differences h. Type of Outlet 5. Unavailable at Time of Application Cost and size of operation were the most common reasons given by vproceseors for not having government inspection. Viewpoint differences were mainly related to the lack of appreciation by government inspectors, for the costs and value of volume in gaining a favorable-cost of pro- duction. Many plant managers felt that they could better manage the quality of their products by maintaining their own quality control personnel. Some plants did not use government inspectors because their customers did not request government certification. Some managers in the packing industry are discontented with the present standards used in grading the processed pack. The majority of the comments received indicated that they thought that the present standards were too low. The most common comments were those relating to the need for drained weight requirements on the cold pack, need for u . Continuous - Inspection covering every phase of processing operations. In-plant - Inspection on only the finished product. 27 a grade B in the standards and the allowance of too many defects in A grade. Many canners believe that government grades do not conform to the trade practice and are too stiff in the top grade. At a meeting of the Michigan Canners Association one cherry packer commented that 'There is no such thing as a fancy grade cherry pack under the standards as set up. It is impossible to segregate sufficient unblemished fruit rj‘ to meet the restrictions. Since there is no intermediate grade, the t result is that practically all cherries become Grade Cfls Many of the processors who complained about the grades did not suggest ways in which they could be improved. Some canners were con- , 3.J cerned with the variations in human Judgment of such things as color and character.16 Several people suggested the use of color charts. Drained weights on cold packs are extremely hard to obtain because of the need _for a uniform temperature and time period for thawing before the check can be taken. 15"Are U.S. Canned Food Grades Commercially Suitable 7' The Food Packer, December 1918, p. 719. 16The research branch of the U.S.Daa. has undertaken research on the measurement and specification of color as a factor of quality with the use of photoelectric instruments. See g§g_Measurement_and. Specification of Color by B. A. Brice. Easthtilisation Research BrQICh, U.S.De‘e CHAPTER IV RACTORS INFBUENCING RAW PRODUCT QUALITY Raw product quality varies considerably between growers, between plants and between days. Many factors contribute to these variations. Many processors attempt to influence grower harvesting and production practices through educational programs and payment plans. Transportation and handling practices affect quality levels. Grower sine and attitudes influence raw product quality. Such non-controllable factors as weather conditions or locational influences are also important. This chapter is an attempt to evaluate these factors. The first part of the chapter is concerned with general compari- sons between plants. Transportation and handling methods, raw product inspection methods, locational influences, payment plans and educational programs are considered in this section. The second part of the chapter consists of a three plant analysis in which seasonal and daily quality variations, grower differences and specific defect differences are computed. Seasonal and daily quality variations are figured and compared for three different plants in different—plante—im-different areas. Controllable and non-controllable defects are related to high, low, and medium quality growers at each of the three plants. Grower size is related to quality at five different plants. The importance of specific defects in.determining raw product quality is correlated between three plants. 29 GENERAL ANALYSIS Transportation and Handlinggnethods Proper transportation methods and careful handling practices are important factors in preserving the Quality of the raw product. The red cherry is a delicate fruit and bruises easily. Unnecessary handling and poor transportation methods should if possible be eliminated and better handling and transportation practices adopted. SOne plants in the state receive most of their raw product from local growers within a 10 mile radius. Other plants receive their raw product from growers over 100 miles away. The distance that must be traveled in order to deliver cherries to a plant has a great influence on the quality that will be delivered. Over 60 per cent of the processing plants interviewed maintained distant receiving stations for cherries. About 75 per cent of these processors stated that they received poorer quality from this source. Cherries are usually transported from the orchard to the receiving station or processing plant in lugs. Since 1952, several plants have adopted the use of water carriers in order to preserve the quality of the raw product during transportation. Dugscald and sunscald are two important defects that have been reduced considerably by this type of transportation. Gaston and Levin stated that processors records showed that the average grade of water transported cherries were significantly 'higher than that received in lugs and that scald and collapsed cherries were eliminated alnost entirely.1 On the other hand, water in these 1Gaston and Levin.' “Transporting Red Cherries in Water from Orchard to Processing Plant," Quarterly Bulletin, Michigan Agricultural Experiment Station, MSU, East Lansing, Michigan, Vol. 37. No. 3. p. 44. Pm“ v Q WMH‘L‘a .3 . 30 tamks must be kept fresh in order to prevent water scald. Without doubt this method of transportation is an improvement over hauling cherries in lugs. This is especially true when the distances hauled increases. The time period from picking of the cherry until it is dumped into a soaking tank is another important quality factor. The average time period for most growers and plants is from four to six hours. The actual range of time estimated by different processors was from one to twenty hours. Those processors that received cherries from indirect sources 2% s.. um‘ I1 . is. (receiving stations) had larger time intervals between picking and tanking. Some growers delivered their cherries to the processor six times a day while others delivered only once a day. Method of Inspection Raw product inspection determines the acceptability of a given load for processing and provides the processor with a basis for making payments to growers. The true quality of the load will be determined from the inspection sample only if this sample is representative of the load. The way in which the inspection and sampling is done, therefore, has an important influence on the quality determination of the load. Proper inspection and sampling procedures are important to processors and growers. ‘Processors do not want to accept low quality fruit that will require extensive sorting to produce a standard grade finished product. Inade- quate sampling and inspection may discourage quality incentive programs 'hy awarding low quality growers premiums and high quality growers dis- counts. The accuracy of sampling the raw product may depend a great deal upon.where the inspectors platform is located. Some inspection stations 31 are located at the tanks and others are located at the scale houses. When the samples are taken from the loads at the scale house, these samples are generally gathered from the top two layers of the load. In many cases this may have encouraged stacking of the boxes and facing the load with better quality cherries. Research conducted by the U.S.D.A. revealed that the percentage of the U.S. NO. 1 quality reported by the official inspector was 5 per cent to 9 per cent higher than that found by the investigators who obtained samples from all parts of the loads as they were being dumped into the soak tanks.2 The inspection of the cherries as they were being dumped enabled the inspector to get a more random and representative sample from the entire load. The size of the sample taken from a load is constant despite the varying size of the loads, for example, inspection records indicated that whether the size of the load is 500 boxes or 5 boxes the sample size remained the same.3 The size of the load will have very little influence on the reliability of a constant size sample if the sample is drawn randomly and the variance of the quality of the different size loads are identical.“ By increasing the size of the sample greater reliability is attained. 2Edward R. Thompson and Raymond L. Spangler. Some Observations on the Relationship of Quality of Fresh Sour Cherries to Thgigglfrqqgsseg Products and Effects of.Processing_of Various Types of Defects, USDA, PMA, Washington, D. 0., June 19149. 9In those cases that appear to be marginal in respect to their acceptance for processing two samples are usually drawn. “Because even very small lots contain several thousand cherries the correctional factor due to the finite size of the universe being sampled is of no practical consequence. ‘ ~ '32 Even though all inspectors may be receiving sour cherries from the same local area, there is often a considerable difference in the average quality graded by each inspector.5 The method of gathering samples and the subjectivity necessary in Judging defects was questioned by many processors and growers. The large volume that is delivered to processors during the peak of the season necessitates quick and fast inspection service. This often results in carelessness in obtaining random samples of growers loads and poor judgement of the quality of the sample drawn. Locati 9311]: Influence: Raw product quality received by plants varied considerably between and Within areas. The average quality received by all cherry processing Plants in Michigan during the 1955 processing season was 92.8 per cent U.S. No. 1 grade. The average quality received by the plants in the northernJcentral and southern areas were 91, 94, and 93 per cent reS‘Dectively. The average quality received by 27 plants are listed on the table on the following W159- \. 5 ts‘ilrrpling of raw product records for inspector differences indicated hat the average quality of raw product sampled by 10 inspectors located a t Traverse City ranged from 5.7% defects to 16.5% defects. \- 33 TABLE III Average Raw Product Quality (per cent U.S. No. 1) Received by 2? Processing Plants in Michigan in 1955 - by area. Northern ‘ A Qentral Southern .___“ 1 90.0 91.7 94.1 2 90.8 93.5 95.2 3 83.5 93.8 93.3 1* 92.2 95.0 92.9 5 93.1 91%? - 9b,.9 6 91.1; 90.9 92.0 7 92.7 9&1 99.6 8 91.7 93.8 91% 9 88.8 93.4 91.2 10 93.1 94.2 95.7 11 99.4 93.9 12 92.7 13 93.8 lb _ 91.7 15 91.7 L...---.__.__,. ...--- __ 91.7 ,_____g m e __l9.9..292}l{_....._----.9.M?-...__.....91.16 .......... .... m 9. . ... __ 1.6 ...._.....J*- --...-__.___..._ Northern area plants received the poorest quality cherries. Varia- ti on in the average quality received by different plants in. this area "as large. Weather conditions had important implications of the raw DPOduct quality. A considerable amount of hail and frost damage occurred in some parts of the area. Dry weather conditions in other sections of t'h‘ls area reduced the size of the fruit. The central area plants received the best quality cherries in 1955. L1 ttle variation existed between plants in relation to the quality re- cOived. Weather conditions were very good in this. area with the exception or the northern part. Southern area plants received fairly good cherries. Those plants that did have poorer quality were concentrated in areas subjected to I.‘ 31+ considerable frost damage and sunscald. Many plants in the southern area received cherries from the central and northern areas. It is, therefore. doubtful whether the average quality received by these plants are representative of the local raw product quality. Payment Programs Processors in Michigan base their payment programs for red cherries .5 a on Michigan Revised or U.S. Standard Grades. Over two-thirds of the processing firms offered the growers incentives for good quality by means of discounts or premiums. Processors recognized the need for . receiving good quality cherries to reduce sorting labor and waste and to maintain quality packs. ‘l'en different types of payment programs are used in the red cherry processing industry. 'I'Velve firms offered no premiums or discounts. Twelve firms offered both premiums and discounts. Keven firms offered discounts only and three firms offered premiums only. The type of paFluent method used and the number of firms using each are as follows: TABLE IV Payment Plans for Raw Product Cherries $18-11 No. Hrms"" .Uescrip‘EI'on' "mayhem“ ---m- ~‘--..—— LO. using A ' 1 10 88 score - 100 score .- 106%‘i?”” .- 2 9 95 score - 100 score - 100% pay. 1% deck for each score below 95. 3 1% for each score above 95, 1% deck for each score below 95. l‘ '3 19% for each score above 95, 1% dock for each score below 95. 5 3 1% for each score above 95, 90—95 score - 100% 1% dock for each score below 90. 6 2 90 - 100 score a 100%, j88-90 score dock 3%. 7 2 Standard price plus % of profits above certain level. 8 1 1/8 cent for cherries 96 or above 1/4 cent for cherries 98 - lOO score. 19 14! per pound above standard price. 0 1 1% for each score above 95. 88-95 score - 100% pay/ \\ v‘ ‘ -. - -.. 35 Premiums and discounts are usually dependent upon the score that the grower received from the raw product inspector. These discounts and premiums are usually stated as a percentage of "the standard price for a pound of cherries. If we assumed that a grower delivered 100 lugs of 98 score cherries each weighing '30 pounds and the processing plant uses payment plan ll» (Table IV) then the grower should receive $309 if 2“ the standard price for cherries is 10»! per pound. _ . : The type of payment plans used by different firms appear to be ,. somewhat related to the competitive conditions existing in the areas in I '3" J which they are located. Similar payment plans are used by plants com- peting for raw product within certain regions. Seven out of eleven plants in the northern area used a dockage payment plan. Six out of eleven plants in the central area offered no premiums or discounts. {Nelve out 01’ nineteen plants located in the southern area offered the growers premiums and discounts or premiums only. ‘me plants in the southern area offered the greatest variety of payment programs to growers. Educational Programs All processors recognized the need for encouraging and helping the g1‘O‘Ier to deliver a better quality cherry to its plants. Fieldmen. grower meetings. mail contacts, and other activities are sponsored by Plants to help fill this need. The larger plants carried on new types 01' educational programs from fieldmen to radio programs. Smaller plants are often quite limited in their educational activities. Mnty-six out of thirty-seven cherry processing plants interviewed had one or more regular fieldmen during the growing season. In other b1”ants the managers work as fieldmsn on apart-time basis when not busy ‘s ’4 \l 36 with packing operations. Large plants employ fieldmen on a year around basis while those plants with a smaller scale of operations employ field- men only during the growing season. Grower contacts made by fieldmen varied from two to six times per grower during the season. Only twelve of 18 processing firms held grower meetings. One firm had .about twelve meetings yearly and another firm held meetings every other year. The district horticultural meetings played important roles in educating the grower in the use of spray materials, fertilizers and grade standards. ‘Ihese meetings are generally held twice a year and are considered by plant managers to be of great importance in grower educa- tion. Mail contacts are the third most popular type of educational activity engaged in by processing plants. Twenty-two out of ‘38 processing firms had one or more mail contacts with growers during the year. Some firms Published monthly newsletters while other firms sent one or two news- letters prior to the harvesting season. Importance of Factors in Determining the _G_1_'ade of Raw Product Average grower size, educational programs, payment plants, and I313nt location of thirty-seven different plants were used as independent Variables to measure their influence on the average raw product delivered ‘9 the plants. The effort to measure the effect of these factors was unsuccessful. Lack of. adequate information for classifying the data and the effect of the necessary aggregation for classification reduced the ac(inracy of the independent variables. The correlations obtained were Very low but it is expected that they would be higher provided more accurate data were available. '3? THREE PLANT AHALYQIS Seasogl and Daily‘Quality Variations. Seasonal movements in raw product quality appeared to be inconsistent from one plant to another. It is normally expected that the seasonal quality movement will be lower in the beginning and at the end of the harvesting season. Plants in the northern area had a greater tendency to follow this typical pattern than did the plants in the south. The seasonal movement of quality delivered to southern area plants was much less pronounced. Many growers start picking their cherries early to reduce the threat of weather uncertainty and to relieve the late season harvesting rush. rI'his practice results in a large amount of undercolored, immature and stell-Lined cherries being delivered to a plant. Late harvesting often I'esults in many cherries that are over-ripe and subjected to additional weather hazards. Pulled pits, scald, decay and bruising defects usually 1“crease late in the season. The comparison of the daily quality received by three plants in 1955 (Figure it) indicated that plant I, located in northern area, had much more day to day quality variation than plant II and III in the sOlltl’iern area. The average quality variation between days in the northern “hint was-1h percent as compared to 10 per cent in the southern plants. The daily quality change exceeded 20 per cent 8 times in the northern Elam; as compared to only 5 times in the two southern plants. The com- 13°31 tion of the specific defects determing the total defects varied during the season. Frost damage, hail, sunscald and other defects related directly to weather conditions played an important part in P wcw>wooom «nan 8 a) Ililtbh- - ,mh I . -pn, - -mW,.+L om ma OH m - «I- .‘ I1-I E v nuompvom HHH ”scam ‘ auonuuom. HH pamflm ..... 1 cuesahoz H uncdm I I I . . mm . . , : f V , z ‘4 V xx, .3 I‘ n - . . . , I s. W , a, . , , 3 , s . d “ _ ~ 7 .om v r ‘ no ¢ E6: . . manuam m I occupruhm> Manacad oozuOAm 3wm kdme 00H I'Wsmi '39 daily quality change at some plants. Quality Differences Among Growers Some specific defects can be controlled by growers through the use of better production, handling and harvesting practices. Color may be improved by proper pruning. Decayed and wormy cherries can be reduced : _ by Drover spray treatments. Pulled pits and. stemmed cherries are P .1 related to picking practices. Lugscald can be reduced by moving the cherries from the orchards to soak tanks as rapidly as possible. Other defects that are considered to be partially controllable are Tr“ v related directly to the weather. Vindwhip, frost damage. hail damage, sunscald and undersize almost always reflect weather conditions. Limb- l”"113. scars and miscellaneous defects may be partially controlled by certain harvesting and production methods but are considered as non- col'ltrollable in this analysis. 'i'ne following charts present a break-down of growers into three dlff’erent groups according to the average percentage of defective fruit delivgred to three plants. Theresare approximately 290 growers at plant I. 50 growers at plant II and 100 growers at plant 111- The high ”3 group represents the one-third of the growers who delivered to the plant the highest quality fruit or the lowest percentage defective fruit. The 1°“ one-third represents those growers who delivered poorer quality which contained a higher percentage of defects. Z40 TABLEV A Grower Differences - Per cent Controllable and Noncontrollable Defects ____ _ Plant I aging/1 Medium $73 Low 1B All Growers Absol. g of Absol. '2 of Absol. 10f Absol {of Defec ts % total % total % total 9!. total defects .defects defects defects defects' defects defects defects ———' Color .42 7.22 . .92 10.10 3. 39 22.45 1.54 15.43 D06” 0 03 o 52 o 1"" 1 0 5h 0 68 L’- 50 0 28 2’ 81 ‘ ‘Horun ; 5“] damage .00 .00 .00 .00 .01 .07 .00 .00 ~ Lug— I scald .51 8.76 .614 7.03 .95 6.29 . .70 7.02 Stems .55 9.45 .65 7.10 2.114 114.17 1.09 10.92 Pulled . . p1 ts .17 2.92 .27 2.96 .22 1.46 .22 2.20 ,J Partially ' " Controll- .9331_e _ _1._68_ g8_._82 __ _ _2._62 28.77 7.36 48.94 3.83 38.38 Vind— --------------- .- - - _ - - whip 1.92 32.81 2.13 23.38 1.87 12.38 1.98 19.84 Scars 1. 16 19. 93 2. 14 23.49 2. 51 16.62 1. 95 19. 5'4 Limb- rub .12 2.06 .26 2.85 .07 .46 .16 1.60 Frost damage .03 .52 .02 .22 .02 .14 .02 .20 3911 .12 1.89 .10 1.10 .11 .72 .11 1.10 scald .45 7.73 .79 8.67 . 1.12 7.42 .79 7. 92 Undser- size .25 4.30 .84 9.22 1.77 11.72 .95 9.52 Birwi Deck . 02 . 34 . 04 . 44 . 07 . 46 . 04 .40 Misc:, .09 1.55 .17 1.86 .17 1.11+ .15 1°50 \_ ...... N°n~Con- trollableh.16 71.13 6.49 71.23 7.71 51.06 6.15 61.62 _\ TOtal we _5.84 100.00 9.11 100.00 15_._10 100.00 9.98 100. 00 141 TABLE V B Grower Differences-Jar Cent Controllable and Noncontrollable Defects Plant II High 1/‘3 Medium 1/‘3 Low 1/3 A11 Growers M Absol. '{ of Absol. i of Absol. i of Absol. % of?“ Defects total % total total total .13-— .. defects defects defects defects defects defects defects defects 1 Color .33 6.18 .33 4.42 .46 4.67 .38 4.85 Decay .01 .19 5 Von ‘ : damage . 01 . 19 . 02 . 20 . 01 . 13 E Lue- scald .33 6.18 .72 9.65 1.05 10.65 .74 9. 1+5 Stems .12 2.25 .17 2.28 .26 2.64 .19 2.43 l w Pulled Dita .02 .37 .02 .27 .12 1.22 .06 .7? Enemy "’ Controll- “w§b§°- - - -'§-2— - 996— - - l-Z‘t— _ $626.2. - _ .1.- 21.. _ 19.39. _. _ 1.28_ __ 17.62 _ n whip 1.66 31. 09 .28' 37. 53 3.76 38.13 2.8? 36-65 311:3 2.36 114.19 2.92 19.714 ’3. 85 39. 05 3.13 39-97 m .- mb .0 . > .0 .64 Frost 3 . 56 . 05 .67 . 06 61 5 Ewee .25 4.68 .32 4.29 .17 1.72 -25 3-19 g 11 .01 .19 . 1m- scald Under- .01 .19 Size ' Bird .19 3.56 .11 1.47 .10 1.01 .13 1.66 "it” .01 .19 .02 .27 .01 .13 g \£'\ .01 .10 .01 .1; f HOn.c° 1' 1 idlabie 4. 52 84.65 6.22 83.38 7.95 80.62 6.45 82.37 AK' To {£1 W 1 f- 5.34 100.00 7.46 100. 00 9.86 100.00 7.83 100.00 \- #2 TABLE V C Grower Differences-Per Cent Controllable and Noncontrollable Defects Plant III High 1/3 Medium 1/3 Low‘ 1/3 111 Growers H- Absol. {of Absol. $1 of 'Absol. 5% of Absol. i of Defects % total 5% total % total % total ‘F— ‘_ defects defects defects defects defects defects defects defects I 'a Color .17 3.31 .36 5.25 .76 7.20 .39 5.45 ‘ ' Ibcay .03 .58 .30 4.37 .38 3.60 .22 ?.0? » Wbrm damage .18 3.52 .09 1.31 021* 2°27 '16 2'23 c , ! Lug— , .1 g scald .08 1.56 .27 3.94 -50 “-7“ '26 7°63 .- 3' Stems .27 5.26 .30 4.37 .60 5.59 .36 5.03 --hwm Pulled bits .28 5.46 .28 4.08 .50 4.74 .34 4.75 FXFtiaJJhy —.——V”O-.' Cantrell- able 1.01 19.69 1.60 23.32 2.98 28.14 1.73 28.16» Wind- -..”. __.__ - - -.-.-._ _ a _ -.. "hit! 1. 57 30.60 1. 94 28. 28 3. 14 29. 76 2. 11 29. 47 gears 1.00 l9.#9 1.52 22.01 .98 18.77 1.44 20.11 inb- . rub .92 17.93 .73 10.64 .60 5.59 .76 10.61 Frost damage .23 4.48 .25 3.64 .33 3.13 .27 3.77 3011 .32 6.24 .21 3.06 .67 6.35 .36 5.03 un— seen :03 , 59 . 42 6. 12 . 83 7. 88 . 38 5- 31 Under- 3129 .02 .39 .07 1.02 ' .02 .19 .04 .56 Bird ‘ NC“ .03 .59 .10 1.46 .02 -28 Misc. ’ .03 .45 .02 .19 .05 -70 \- -.. “m... — ...-‘ __ - ..-- ..., ._ .... _ o ‘— lion"(3021. _tr°11&b1e 4.12 80.31 5.27 76.68 6.59 71.86 5.43 75.84 Ave '— ..‘Trflmfl .... —-- “"‘“ ' tal Dgf' 5.11 100.00 6.87 100.00 9.57 100.00 7.16 100.00 . \._ ~ ~ “- - - ‘- w 4.3 The previous tables proved invaluable in making comparisons between growers at a plant. At plant I abouta 10 per cent difference exists in the amount of controllable defects from the high quality to the low quality groups. At plant II this difference amounted to 2 per cent. At plant III the difference was 8.5 per cent. These figures infer that growers who deliver high quality attempt to remove those defects that J_____ g ‘i are considered. to be partially controllable. Size of Grower ‘ The quantity delivered by the individual grower appeares to have an important influence on the quality of the raw product that is delivered by the growers to the plant. Extremely large quantity pro- ducers are unable to directly oversee the picking operations and are DrObably unable to get their cherry crop harvested at the best time. Average size growers are better able to see that the picking operations are doneproperly and that the crop is harvested at the right time. S"filler growers probably lacked the proper equipment to take care of the 'Droduction and harvesting of the crop. These probable assumptions 3“ borne out by the following graph: A‘verage Percent Defects - by Grower Size - By Plant FIGURE 5 5 Defects- f Plant I .... Plant II --- It” , Plant III-v" Plant I? as ’1- Plant l-t-IJ 1 - 1 12 . I M I ll 5 10 J 1 9 q . h. “1 *.+\.'e 8 d 4‘43"" ‘4- ‘+ 7 u N ~ ' ‘~ ' -e n \ 7" 6 {1" 7: d 8 / 5 q g . 8 8 8 8 8 8 8 o o o o o m H o. o. u: N. 0-: .-+ H I I I‘ I I I o o o o o o o o o o o o o o. n H u: ...? .1 Size of Grower (Number of Lugs) 45 Two hundred growers from plant I in the northern area were compared with two hundred growers from plant II and III in the southern area. This comparison revealed that the range of grower size and quality delivered by the growers was much wider at the northern plant. The size of the growers as determined by the number of boxes delivered to the plant ranged from less than 1,000 to over 15, 000 at the northern plant and from less than 1000 to only 9,000 boxes at the southern plants. The average per cent defects delivered by these growers ranged from 2 to 26 at the northern plant and from 2 to 15 at the southern plants. Importance of Specific Defe_cts in Determinigg 333’. Product Qualit} The raw product inspection records of three plants were analyzed to determine the difference in the relative importance of specific defec ts at different plants. Plant I is located in the northern area and plants II and III are located in the southern area. The results of this analysis are recorded in the following table. 46 TABLE VI Relative Importance of Specific Defects - 3 plants6 fleets Plant I Plant II Plant III findwhip 1 1 2 Sears 2 2 1 Color '1 1+ 1+ Stems 1+ 6 6 Undersize 5 14 7 ri— Sunecald 6 5 13 ' Lugscald '7 I 10 7. 3 Decaor 8 11 14 Pulled pits 9 . , 8 8 i Limbrmm 10. ~ 1’ 3 fl-‘ 9 I Miscellaneous 11 ‘r' 13 15 Hail damage 12 ’1‘ 7 10 Bird pack 13 . 15 12 . Frost damage 115 (.7) 9 «‘3 ' 5 .Ieua_éksmuni_.__, 15 - 4. 12 1~ 11 The rank order correlations between plants I. II, and III shows that vsignificant dependence in ranking exists. ‘Ihe rank order corre- lation of incoming defects between two plants indicate that there is a Closer agreement in the ranking of the northern plant to each of the southern plants than there is between the two southern plants. BirdpeCk. miscellaneous and worm damage are of minor importance while windwhip. scars. and color are very important. There is considerable difference in the ranking of the defects which are of minor importance. M 6 R I. 11. III = .582 : (Significant at 5% level) I. II = .601, . (Significant at 1% level) R I. III = .585 I (Significant at 5% 1eve1)~ R II. III = .557 = (Significant at 53 level) CHAPTER V INPLANT QUALITY FACTORS AND QUALITY RELATIONSHIPS In the" last chapter the factors influencing raw product quality were discussed. Though satisfactory measurement of differences between plants is not feasible it is probable that weather conditions, payment plans. 1 educational programs, and grower characteristics will all influence r—— quality to some extent. The analysis of three plants indicated that a ‘1 REE“ highly variable daily quality pattern existed. Quality differences between groups of growers revealed that good quality growers had a lower amount of controllable defects in their cherries. Large size and small size growers usually had poorer quality fruit. Windwhip. scars, and lack of color were important defects affecting raw product quality. Raw product quality is but one of many other factors influencing the finished product quality. After the fruit has been delivered to the plant. quality of the fruit may be lowered, maintained, or improvedby processing oI’C’l‘atlons. This chapter will discuss these in—plant operations and “3111?? factors. Handling methods, inspection methods, sorting methods and Other processing operations can influence quality. Quality relation- Ships are discussed in the second part of this chapter. An understanding °f these relationships will be useful as aids for the evaluation of gambling for quality control. III-PLANT QUALITY FACTORS Handling MeLhods in Receiving cherries should be transferred from lug boxes or water tank trucks with the minimum amount of bruising. Consideration should be made by the 1*? 1&8 processor as to which method of receiving cherries will cause a minimum amount of damage. It the present time many plants dump cherries directly into the soak tanks while others dump cherries into booted water conveyors or into hoppers which are carried by belt conveyors above the soak tanks where they are emptied. Bruised cherries have important consequences in determining the yield and quality of the finished product. Research conducted by Hills and Whittenberger revealed that unbruised red cherries fit“ : soaked in water increased in weight and firmness and decreased in soluble— solids content and acidity while bruised cherries did not gain significantly _l in weight and lost appreciable quantities of soluble solids and acids.1 4' L e. The soak tanks not only wash the cherries and provide a medium for storage but they also‘ provide a means by which the cherries can be firmed {01‘ processing operations. A reduction in the temperature at which the cherries are stored either in air or in water, increases the firmness of both bruised and unbruised cherries.2 Ih‘. V. F. Robertson Of Michigan State University stated that when the temperature in the center of the chem? is the same as the temperature of the medium in which it is stored additional periods of storage will not increase the firmness of the cherry.3 5 positive relationship exists between the length of soaking time and the temperature of the water in the soak tanks amongst 15110 37 plants sampled on this study. \ 1 (Sande H. Hills and R. T. Whittenberger, Studies on the Processing of ed Cherries, Food Technology". 1953. V01. VII. N0. 1. p. 31. 2mm, 3 P01‘somii interview with Dr. H. r. Robertson of the Horticulture Depart- “fit at Michigan State University. 49 Very few processors keep cherries in soaking tanks over 2b 'hours. Extremely long firming periods may cause cherries to split while short. firming periods may not allow sufficient firming for good processing. Results obtained from soaking cherries from zero to 1+8 hours in water supplied continuously at 145° F. and 37° F. indicated that water at lower temperatures is more desirable and that the firming period should be limited to twelve hours or less to reduce the loss of soluble solid, and color and to minimize the amount of cullage.l+ Northern plants have a valuable asset in being able to obtain a cheap and abundant supply of water below 50°F. The average temperatures in the firming tanks are 52. 1’8. and “-50 I". in the southern, central and northern area plants respec- tively, iorting Hoses“ Cherries are flumed from the soaking tanks into conveyors that carry the fruit over eliminators. As the fruit passes over the eliminators, undersized and cull cherries drop between the horizontal bars while the largfil‘ cherries continue toward the inspection 1591,109- About 1% 0f “1° “”11 volume of cherries received by the plants are removed from the processing line in this operation. For a plant processing 3130‘“ 10.000 ”mud“! of cherries in an hour this piece of equipment 18 worth 8-‘0'01‘011' ““91? 5 sorters who can each remove 20 pounds of undersized cherries in a“ hour . Eliminators are not only used for removing undersized cull cherries but they are also used as grade sizers for good cherries. The factor of M. 4, R0? B. Marshall. Cherries .andHQhergy Products, Interscience Publishers, 1%.. New York, 1952+, 'p. 196. 50 Inuform size is important in the determination of pack quality. By the use of size graders this uniformity can be obtained in a very cheap and economical manner. Undersized cherries normally are processed for special purposes or used for cherry Juice. Processors should maintain two or more eliminators in the processing lines to remove undersized cherries and foreign materials. Eliminators are efficient equipment for reducing sorting labor requirements. As grade sizers they assure the processor of getting a more uniform size in his tack. Pie-making concerns buy much of the undersized fruit. Those cherries that pass over the eliminators usually will pass Grade A require- ments for size and uniformity. Cherries are guided from the distributing belts on to inspection or sorting tables. The average sorting table is twelve to fourteen feet in length. Most processors have one sorter for every two linear feet of belt. Some plants have the sorters working close together while others have the sorters spread along the table as much as possible., Many plants have sorters stationed along the distributing belt and between the pitters. This is frequently the case in plants which do not have an adequate number or large enough sorting tables. Wide differences existed in the number of sorters that processors maintained on their line. One processor stated that he employed only enough sorters for the purpose of keeping the government officials happy. Other processors employed large numbers of sorters with the objective of improving quality. The number of sorters per thousand pounds of cherries going through the processing lines in an hour ranged from 1.30 to 6.6? among thirty-seven different plants. Adjustments in the rate of’floy per sorter are often made by e- _ - 51 reducing the volume or by reducing the speed of the belts. Most plants nude adjustments by changing the volume going over the belts while nmintaining a belt speed of around twenty feet per minute. The number of sorters maintained by the plants varied little from day to day regardless of the quality. Unexpected weather hazards may reduce the incoming quality rapidly and most processors saw the need for holding a constant crew for such conditions. Time studies in the processing plants revealed that considerable differences exist between sorters in the quantity of cherries that are sorted out in an hour.5 Studies conducted in twenty-six processing plants on 155 sorters showed a variation of from 5.5 to #2.b pounds of cherries in the hourly rate of pick.6 Plants receiving poorer quality generally had the higher pick out. The efficiency of the sorter seemed to be reduced in those plants that were visited in the afternoon. This can probably be explained by the increased fatigue among the workers. The range of the percentage pick out among plants is large, varying from‘9.6% to 2%. Most managers.stated that sorters on belts could not sort out more than 5%. The average for all plants as determined by the time studies is 4.6%. The differences in the percentage pick outs may by attributed mainly to different quality received and different quality standards set by the plant management, though both the time of the observation, and the size of the cherries may influence the performance of the workers. -..— 5These time studies were conducted on each sorter for 5 minutes. The count was then multiplied by 11 to arrive at an hourly figure that allowed for a 5 minute break per hour. 6 This pick out was figured by calculating the average number of cherries sorted out per hour divided by 120. 52 .l .J m.o mom Hm.m ma.ma Hm.mm N¢.HH m mm os.e .mm ao.m or om.: .aw m.a com ma.a No.mm mm.wm m~.oa m cm mmm.oa .mm 00.: m: oo.~H .NN a.m ado mo.~ NH.©~ mm.mm ea.aa m mm oo.ea .HN -.~ . . om oo.m .om «.3 mmm oo.~ am.am ~a.~m a~.o~ a «a oo.w ‘ .oH o.m mam mm.~ ma.oa ma.ma m~.m a an oo.o~ .wn s.m 5mm aa.~ :m.aa aa.o~ mm.e : mm am.HH .aa _ and m 2.6 .3 aa.m ma om.m .mH e.: mm: oo.~ ~m.- om.am rm.ma a o~ oo.oH .aa a.m mom ma.a mo.am Hm.NN m~.aa m a” oo.m .ma mm.~ mm oo.m .NH oo.~ . NH oo.o .HH mm.~ mm om.w .oH a.~ om: om.~ mm.oa Ho.ea ma.ma . a on oo.o~ .a a.~ sow mm.m mm.ma Hm.H~ om.m w as oo.m~ .m m: oo.ma .a o.~ mma on.” .oa.ma mm.am . aH.m a ma oo.m .e m.a flown m~.~ em.mm as.~a a~.HN a or oo.wa .m m.m was aa.m Hm.H~ am.~m No.ma m a« oo.a .a 3.0 send mm.m mm.m~ oa.mm om.m oH an oc.em .m e.m mmm mo.m mm.wa we.sm mo.oa m on am.« .N a.m was m~.« so.aa mm.ma am.m : mm oo.NH .H 1 25 6.: .92 gm :3 need! €545 .o.m cabana oooa pop .eoea» stepson outspom Anna oooav unsam .280 new eunuch 2.93:5 spouaom pop 30 £03 no ewfia no no ..5 non H6009 uhovhom Wop-n52 hang ogHob I A! ‘1‘” [HI n 3:on wnamnoooum an I moagpomaom wagaom HHb M92 3| 53 ll.‘.i‘ at ...—r—W .hano moan» aaouaom Bonn confisaouoc owner .aaoaaom was you ommno>m on» ma mesa» owoaobm on» can» wcuapnmd Any .pH H u noaaaogo oma an“: season an asoaoap ac omaum Asa m.m mam mm.~ on.sa oa.ma aa.m m «a om.m .am om.a Na oo.wa .mm o.m aam ma.~ mm.ma H~.mm aw.aa a ma on.wa .mm H.a NNHH mm.a oa.ma om.a~ aa.o m N: oo.a~ .am m.a ”Hm am.~ n~.oa oa.oa mm.aa m on oo.aa .nr m.~ coo om.~ om.aa mm.om am.w NH om oo.am .mm o.m mama ~m.a mm.aa mm.am ma.m~ m Ha oo.om .Hm a.m new no.” ma.aa mm.- ao.oa w mu oo.ma .om ow oo.m~ .mm m.m ma: -.~ aa.m~ m~.om ~:.aa o ow oo.m .mm ~.a mom ~:.m mm.~a so.ma ma.a a a: oo.~a .am a.: mom m~.m ma.aa N¢.NN ma.a m or oo.ma .om .o.m use .upa .o>< swam you soars eoaaansm anoo Rom dugoum 80H hog .erg wheat—om whoa-«om Ann." OOOHV ucddm eunuch mohoaain pooaom you one Moan no omccm no no .un non Hdvon. uhuahow hoe—52 hoe—BB DEE—”ob A.v:oov HH> MHmiH 50 TKBLE VIII Cost of Sorting Labor per Thousand Pounds'V/mr: at Different Volume Levels and Different number of sorters Volume/hr. 100 80 60 00 20 10 1000 lbs. Sorters Sorters Sorters Sorters sorters Sorters 50 $ 2.00 . S 1.60 $ 1.20 $ .80 $ .00 $ .20 05 2.22 1.78 1.33 .89 .00 .22 00 2.50 2.00 1.50 1.00 .50 .25 35 2.86 2.29 1.71 1.10 .57 .29 30 3.33 2.67 2.17 1.33 .67 .33 25 0. 00 3.20 2.00 1.60 .80 .00 20 5.00 0. 00 3. 00 2.00 1.00 .50 18 5.55 0.00 3.33 2.22 1.11 .55 16 6.25 5.00 3.75 2.50 1.25 .62 10 7.10 5.71 0.29 2.86 1.03 .71 12 8.33 6.67 5.00 3.33 1.67 -83 10 10.00 8.00 6.00 0.00 2.00 1.06 9 11.11 8.89 6.67 0.00 2.22 1.11 8 12.50 10.00 7.50 5.00 2.50 1.25 7 10.29 11.03 8.57 5.71 2.86 1.03 6 16.66 13.33 10.00 6.67 3.33 1.66 5 20.00 16.00 12.00 8.00 0.00 2.00 0 25:00 20.00 15.00 10.00 5.00 2.50 3 33,s3 26.67 20.00 13.33 6.67 3.33 2 50.00 00.00 30.00 20.00 10.00 5.00 1 100.00 80.00 60.00 00. 00 - __2_g._ 00 0.99_ ‘Based on wage rate of $1.00 per hour. 55 TKBLE IX Cost of Sorting Labor per Thousand Poundswm at Different volume Levels and Different Number of Sorters =—-O-——'-. m ‘-m -‘ Volume/hr. 100 80 6 00 20 10 1000 lbs. Sorters Sorters Sorters Sbrters 'Sbrters Sorters 50 $ 1.70 S 1.36 $ 1.02 S .68 $ .30 $ .17 05 1.89 1.51 1.13 .76 .37 .19 00 2.13 1.70 1.28 .85 .02 .21 35 2.03 1.95 1.05 .97 .08 .25 30 2.83 2.27 1.80 1.13 .57 .28 25 3.00 2.72 2.00 1.36 .68 .30 20 0.25 3.00 2.55 1.70 .85 .02 18 0.72 3.77 2.83 1.89 .90 .07 16 5.31 0.25 3.19 2.12 1.06 .53 10 6.07 0.85 3.65 2.03 1.22 .60 12 7.08 5.67 0.25 - 2.83 1.02 .71 10 8.50 6.80 5.10 3.00 1.70 .85 9 9.00 7.56 5.67 3.77 1.89 .90 8 10.62 8.50 6.38 0.25 2.12 1.06 7 12.15 9.72 7.28 0.85 2.03 1.22 6 10.16 11.33 8.50 5.67 2.83 1.01 5 17.00 13.60 10.20 6.80 3.00 1.70 0 21.25 17.00 12.75 8.50 0.25 2.12 3 28.33 22.67 17.00 11.33 5.67 2.83 2 02.50 30.00 25.50 17.00 8.50 0.25 1 85.00 68 00 51.00 30.00 17.00 8.50 .—_‘ ‘ _ ...—..4 j _ O.“ W —— ‘Based on wage rate of 85¢ hour. 56 Scrting labor is an important part of the labor costs incurred in the processing business. Quality adjustments are made by reducing or increasing the volume going by a given number of sorters on the inspection tables in a given time period, or by increasing or decreasing the number of sorters on the inspection tables. Tables VIII and Ix show the costs twr thousand pounds of cherries for different volumes passing over inspection belts for different numbers of sorters. Table X indicates the number of sorters needed to upgrade fruit of varying quality and of different rates of volume. TKBLBIX Man Hours of Sbrting Labor Required to Up Grade Fruit Using Average Pickout Rate' 133786Er “ ' """""""“'"”"""' over Percent Defective Fruit to be Removed sorting .1...-_..._2 3 “E 5 3“"7m‘8 9 ‘10 .11 12 belt ‘— A .4 — 50,000 25 50 75 100 125 150 175 200 225 250 275 300 05,000 22.5 05 67.5 90 112.5 135 157.5 180 202.5 225 207.5 270 00,000 20 00 60 80 100 120 100 160 180 200 220 200 351000 17.5 35 52.5 70 87.5 105 122.5 100 157.5 175 192.5 210 30,000 15 30 05 60 75 90 105 120 135 150 165 180 25,000 12.5 25 37.5 50 62.5 75 87.5 100 112.5 125 137.5 150 20,000 10 20 30 00 50 60 70 80 90 100 110 120 18,000 9 18 27 36 05 50 63 72 81 90 99 108 16,000 8 16 20 32 00 08 56 60 72 80 88 .96 10,000 7 10 21 28 35 02 09 56 63 70 77 80 12,000 6 12 18 20 3o 36 02 08 50 60 66 72 10,000 5 10 15 20 25 30 35 00 05 50 55 60 9,000 .0.5 9 13.5 18 22.5 27 31.5 36 00.5 05 09.5 50 8,000 0 8 12 16 20 20 28 32 36 00 00 08 7,000 3.5 7 10.5 10 17.5 21 20.5 28 31.5 35 38.5 02 6,000 3 6 9 12 15 18 21 20 27 30 33 36 5,000 2.5 5 7.5 10 12.5 15 17.5 20 22.5 25 27.5 30 0.000 2 0 6 8 10 12 10 16 18 20 22 20 3,000 1.5 3 0.5 6 7.5 9 10.5 12 13.5 15 16.5 18 $2,000 1 2 3 0 5 6 7 8 9 10 11 12 1,000 .5 1 1.5 2 2.5 3 3.5 0 0.5 5 5.5 6 . ‘-OO‘— Igguming that the average sorter can remove 201s of’defectivefifruit in an hour. 5? Variations in rates of pick out of different sorters timed in different plants indicates that no one rule may be applied to sorting labor requirements. As the quality of the incoming fruit improves it becomes more difficult to detect the defective fruit. Quality of the incoming fruit and the efficiency of the sorting labor are variables that have important implications in determination of the sorting labor require- ments. These variables must be considered in using Table X. Many processors are selective in hiring sorters to work on the tables. One plant gave prospective employees a marble test. These marbles were colored and with blemished spots. Each participant had to sort out the undercolored and blemished marbles within a given time period. The manager said that this test proved to be very useful in preventing extremely slow sorters and colorblind pe0p1e from working on their sorting tables. Processors are not only faced with the problem of insufficient amounts of defective cherries being removed by the sorters on the sorting tables but they also have the problem of too many good cherries being removed. Inaccurate sorting frequently results in many good cherries being discarded in the pick-outs.' One plant inspector has checked the Dick outs of certain individuals on the sorting tables and has often found that these pick—outs included over 50% absolutely good cherries. Florescent lighting is provided by all processors over the sorting tables. Two plants had colored lighting in order for their workers to better detect the small but major defects in the cherry. An experimenta- tion on the use of colored lights revealed that blue florescent lights increased sorting efficiency 10 to 30% and that red florescent lights increased sorting efficiency about 60% over the ordinary white g-h‘l - 1 ~ w-“T'L I 'W 58 florescent light.7 Most cherries are rolled around by hand in order to detect defects on all sides of the cherry. The use of foam rubber fingers placed over an inspection belt proved 80% effective in turning the cherries over for a.m0re complete inspection.8 Much of the sorters time is utilized in rolling the cherries over instead of picking cut the defects. The defects on a cherry are extremely hard to find because of the tendency for the defect to have a flattening effect on the side of the cherry. This usually results in having the defect face down on the belt with its mobility reduced. All plants have some person supervising the inspection belts to co- ordinate the flow of the cherries with the quality and the rate of sorting. Larger plants have head floor-ladies in charge of all inspection belts and sub-floor ladies in charge of each line. Usually a line consists of one to four inspection tables._ Sub-flcor ladies generally work near the end of the sorting belts where they can judge the quality going through the lines. Fitting Operation After passing over the inspection tables the cherries are fed into an automatic pitting machine. The pitting operation has important quality 7G. M. Peterson and W. M. Carleton. Possibilities of Using Colored Lights for Detection of Cull Cherries. Quarterly Bulletin, Michigan Agricultural Experiment Station, East Lansing, Michigan, V01. 30, No. 2, Nov. 1951, pp. 177-178. R 'Ibid., p. 180, 59 hmflioations that are related to the firming process. Cherries that are xuw firmed properly will tear and lose more Juice during pitting. Also many surface blemishes have a chance of being removed by the needle plunger. The pitters generally used for sour cherries have the capacity of one ton of cherries per hour. Some processors attempt to operate pitters above this capacity during the peak of the season with the result that some cherries with pits get through the pitters. Some processors main- .«wue tained sorters between pitters to prevent cherries with pits from flowing f' 'il' .' through. Most processors operate pitters at 85% of the rated capacity. The "" "“* number of pitters in use varied from ?8 in the largest plant to l in the smallest plant. Most processors maintain four to six pitters per line. The check maintained on pitter operation varied from one plant to another. The amount of pitter supervision varied from one man in charge of 18 pitters in one plant to one man in charge of 6 pitters in another plant. Close supervision is necessary when pitters are operating near capacity. _flptpack Process V troughs and circular semi-automatic hand pack fillers are commonly used for filling cherry containers. The circular filler is a revolving stainless steel table with circular holes around the edges under which the container is located. The Y trough consists of a trough with a gate across the end. Containers are automatically positioned under this trough by means of a conveyor. The V trough is used mostly for No. 10's while No. 2 and No. s03 are usually filled by circular 60 handpack. After filling the containers it is customary to check the fill weights of the cans. Exact weight scales are generally situated near the line and cans are removed as they are filled. All No. 10 cans are usually checked on the fill weight. The spot check On the smaller con- tainers varies between plants from that of every fifth can to that of every 10,000th can. water or syrup is added to the contents of the container before it goes into the exhaust box. The great majority of red cherries are packed as pie stock and are therefore canned in water. In recent years there has been an increase in the syrup packing of red cherries in an attempt to create new consumer uses for the product. After the cans are filled they are conveyed into exhaust boxes to produce a vacuum in the can when it is closed. Open cans are fed onto a wide metal belt that conveys the cans through a shallow tank of hot water for a certain period of time. Gear type exhaust boxes are also used in which these cans are conveyed back and forth throughout the exhaust box in single file. Most processors attempt to obtain a 160 degree temperature in the center of the container. From the exhaust box the cans are conveyed on a belt to a closing machine, that seals the cover on the can. Most processors run the No. 10 closing machine at about 160 cans per mihute. The smaller cans are run through closing machines at about 200 cans per minute. These machines are located close to the exhaust boxes so that the vacuum created by the exhaust boxes will not be lost. Red cherries are processed in retorts or in continuous cookers. Most continuous cookers and coolers are of the flat type constructed 61 shfilar to the exhaust boxes. Four of the plants visited have reel type mumprs and coolers that conserved considerable plant floor space. 'h-renty—four have continuous cookers and coolers and ten plants have retorts. Retorts enabled the processors to better adjust his cooking time flu-different products. The cans in the retorts are cooled by passing the cages through tanks of cool water. Most processors attempt to obtain a temperature of 200 degrees F. in the center of the cans during cooking and 100° F. in the center of the cans after cooking. Cold Pack Process ya—s=»-_. j After the pitting operations cherries to be coldpacked are generally putinto 30, 15, or 10 lb. tins. Out of twenty—five plants that coldpack, eighteen plants use a trough or hopper type filler with a sliding gate and the seven plants have automatic fillers. Red cherries are usually cold- tmcked with sugar added to its contents. The most common type of mixture for 30 lb. tins are 25 lb. of cherries and 5 lb. of sugar. .Sugar is usually added with a measured scoop for hand filling operations. The automatic filler usually adds the sugar automatically as the container is being filled. All large cold pack cherry tins are weighed on exact weight scales. The automatic fillers not only add cherries and sugar but also weighs and stamps the contents as the container is being filled. The filling operation automatically stops when the scale is pressed down. The hand filling operations consist of adding cherries and sugar to the containers contents until the desired reading is obtained on the scales. Inspection Method; .The thoroughness of the check on the quality being packed varies 62 frmnone plant to another. Plants that have continuous government inspection service have rather complete and systematic checks on the mwflity flowing through the plant. Those plants which do not have this service vary widely in their checking methods. Checks are made by plant mmfloyees from randomly to every 15 minutes. The persons obtaining this check ranged all the way from specialized quality control personnel to floor ladies. The inspection of the product before processing is done according to the standards for the raw product while the inspection on the finished tmoduct is done according to the U. 3. Standards for processed cherries. Practically all plants check the drained weights, the fill weight; and the net weights and the vacuum, similarly as it is done by government inspectors. About one-half of the plants without governmental inspection service maintain records of their inspection. Government inspectors determine a probable grade and a final grade on.the finished product. The probable grade is determined the day the product is processed and the final grade is determined after the product has been stored for a period of time. changes in the product occur after processing and usually the final grade will be higher than than the probable grade due to the blending of color and other factors. The final grade on the coldpack is usually determined one or two days after processing but the hot pack is usually allowed to cure in storage for ten days before a final grade is determined. QUALITY RELATIONSHIPS Raw and Final Product Quality An attempt was made to correlate the raw product grade to the final 63 pnmuct score. The average proportion of fresh cherries that graded U. S. No.1.for the current and the previous day was correlated with the average score of the finished product for the current day. This resulted in two significant correlations at two plants and no significant relationship at the third plant. It is doubtful whether the grade designations on the fresh cherries trove to be a satisfactory index of the quality of the processed product. Spengler and Thompson noted that some ofthe defects which are scored in grading the fresh cherries are of little consequence in the grading of the canned or frozen product.9 Their study indicated that if the fresh cherries are of such a nature that they do not seriously affect processing A quality, they can easily obtain a much higher grade in the processed inspection. Cherries with attached stems are easily removed by sorters and undersize cherries are removed by eliminators. Qcald and off-colored cherries usually blend with the other cherries and do not present much of a, prdblem if the defects are not too serious. The correlations indicate that quality is reduced as it is processed. At one of the plants the mean value of the raw product is 91.b2%‘U. S. no. l grade cherries while the average total scere on the processed product is 85.72%. This problem is not one of quality deteriation but simply one of dissimilarity of standards used in evaluating the raw and processed grades. _‘ M Thompson and Spengler. 'MflW—W Q Influence of Raw Product Quality on Finished Product Score Finished Product Score Finished Product Score FIGURE 6A Plant I A V: 76.87 9 .1qu 7' .309* e L j ——1 fl r v 1 1 T —f W I I 1 T 1 81 82 83 8b 85 86 87 88 89 9o 91 92 93 9h 95 % U.S. No. l - Raw Product Won-significant at the 5% level ‘PIGURE 63 Plant II A y= 26.37 + .6921: V I 563. . f ‘—' ‘V fi—w ‘— T U V 81 82 83 8h 85 86 8? 88 89 9o 91 92 93 94 95 $ U.S. so. 1 - Raw Product ‘ Significant at the 3% level 64 fins-rs Finished Product Score 95. 9“. 91. 9o. 89. 88. 87' 85. 84. 83. 82J 81W 80 65 FIGURE 60 Plant III ‘jie 23.25 . .669: r 3 .1433. 818:2838u83828%888§939i9&939L55 9E U.S. No. 1 - Raw Product ‘Significant at the 5% level 66 Certain defects in the raw product are very important in determining unequality of the processed product. Plant managers were asked to rank these defects in their order of importance in affecting the grade of the trooessed product. The following chart illustrates the points system used in evaluating the importance of these defects and their effect on ——-v] the hot, cold and total pack. ' . TABLE XI Importance of Raw Product Defects in Determining Processed Prod- E j duct Quality at 3? Plants " .- -—- Scoring System Total Pack Hot Pack Cold Pack . Points Rel. Points Rel. Points Rel. Fumk order Points Defect 1 7 windwhip 318 35% 173 39% 145 30% 2 6 Lugscald 198 21 68 15 126 26 '3 5 Cblor 180 19 75 17 , 105 22 h b Undersize 106 ll 59 13 ' #7 10 5 3 Sunscald 63' 7 39 9 24 5 6 2 Character 2h 3 12 3 12 3 7 1 Hall 17 2 10 2 7' 2 ' Maturity 12 1 6 1 6 1 .___ _ Stems 10 1 4 1 6 1 924 100 ##6 100 478 ‘100 Vindwhip, lugscald and color are the most important factors affecting the processed product quality. A rank order correlation of the defects affecting the hot and cold packs resulted in an R value of .985. This is very high because the only difference in the rankings of the defects is a switch in the relative importance of color and lugscalds Lugscald is more evident in the cold pack because of the absence of a cooking process to blend the color throughout the contents. Maturity and char- acter of the cherry are not listed as defects in the raw product as such, 67 but usually are related to the sugar content and the firmness of the cherry. There is some difference between the importance of raw product defects in affecting the processed product quality and the importance of specific defects in determining the raw product grade. Scars are not mentioned by the plant managers as defects affecting their processed product. It could probably be assumed that they considered windwhip and scars as synonomous terms even though there is a distinct difference between these defects. Lugscald is comparatively important in the processed product but is ranked fairly low as a factor determining the raw product grade in two of the plants. Those defects not ranked as factors affecting the processed product are low in the rank of importance of the specific defects in determining the raw product grade. Importance of Factors Determining Final Product Score Color, defects and character are the three quality factors that ‘ determine the final product scores. Four hundred inspection samples of the processed product scores were analyzed on one of the plants. The effects and importance of each factor on the total scores is included in the following charts. Defects accounted for 39.2 per cent of the differences and character and color accounted for 35.3 and 25.3 per cent of the differences respectively. Two hundred samples of hotpack inspections showed that defects are the most important factor in determining the total score of the hotpack. Defects, character and color accounted for 43.9. 32.5, and 23.6 per cent of the differences respectively between the potential and the actual count on the scores. The coldpack inspection samples revealed that 68 clnaracter is the most important factor affecting the total score. Defects, character and color accounted for 31+, 38.5, and 27.5 per cent of the differences respectively between the possible and the actual scores on the coldpack. . TABLE III Relative Importance of Color, Defects and Character in Determining Finished Product Quality (400 samples Hot and Cold Packs) Potential Actual Count Difference Factor _____...._..----....._._. --__.._.-._..--- -- ‘ Points Per cent Points Per cent '-Absol.¥r- iTotal Color 8,000 20 6970 17.0 2.6 25.5 Defect 16,000 00 10397 36.0 0.0 29.2 Character 16,000 00 14552 36.4 3.6 35.3 Total Score 00,000 100% 35919 89.8% 10.2% 100.0% Hotpack (200 samples) Color 0,000 20 3065 17.3 2.7 23.6 Defects ' 8,000 40 7006 35.0 5.0 43.9 Character 8,000 00 7258 36.3 3.7 32.5 Total Score 20,000 100% 17729 88.6% 11.4% 100.0% Coldpack (200 samples) Color ' u,000 20 3505 17.5 2.5 27.5 Ihfects 8,000 40 7391 36.9 3.1 34.0 Character 8,000 #0 7290 36.5 3.5 38.5 Total score 20,000 100% 18190 90.0% 9.1% 100.0% “* ~--“ ‘M-----C-“ ‘dMHO ' “ma-~o—‘ ‘- Fifteen per cent of the 400 samples graded C or standard grade. kaects caused an samples to grade standard. Color caused 12 of the samples 69 to grade standard and character was responsible for placing four samples in grade C. Color has much less influence on the total score but it can affect the grade more easily because it allows for the dockage of only three points before the sample falls into a lower grade. Defects and character are each allowed six points before the grade is dropped. ihtal score cannot affect the grade unless one of the three factors is below the minimum number of points. K multiple correlation of the three factors that determine the total score for 200 samples grouped into fours indicated the possible influence that defects had on affecting the total score. The following R.values were obtained in attempting to get the interaction of these four factors on each other.10 m- -----~om”. Elly-'3 .335 12:7508 R: .999 szy = .933 Rxlx3 = .499 Rxay = .523 szx3 = .217 m«--”-ooo‘-.~~—-‘Ofi-m-~" C The above R values indicated that defects and character are more important as factors determining the processed score. The intercorrela- tion between color and character is much higher than that between color and defects or character and defects. These results were expected on basis of the preliminary exam of the records. Defects had greater variation than the other two factors and was therefore, expected to have greater influence on the total scores. Color and character were expected to be more highly correlated because of the relationship of immaturity and scald in affecting both character and color. ~O. ..m““ 10 These factors are 11 a color; 1 total score. = defects, x = character, and Y = 2 3 70 ...,LLM 011531.621 .111 .Einlaaeimdtrmmi Processed product quality changed throughout the season. At one large plant defect and character were important factors in decreasing product quality as the season progressed. Color improved the grade of the finished product as the season progressed. Most plants have better quality in their coldpack than in their hotpack. This was either because the plant attempted to pack better quality in the coldpack or the defects contained in the coldpack are not as noticeable to inspectors. Defects and character in the coldpack are considerably lower than in the hotpack. Color improved daily in the coldpack while in the hotpack the color remained about constant. Drained'Weightg 'Drained weights are not affected either by soluble solids content or by time and temperature of soak, but varied with relation to areas and years and with weather conditions. Small amounts of sugar, dry or syrup, gave higher drained weights and better quality without increasing costs)11 The above statement condenses the results of a five-year study by Bedford and Robertson on the factors affecting drained weights on sour cherries. The tissue structure of the cherry that was formed three or four weeks before harvest appeared to be the significant factor influencing drained weights. Bedford stated that drained weights on cherries were probably influenced by three things: (1) weather conditions, (2) nitro- genious fertiliser, and (3) harvest maturity.12 W‘Q‘H 11Bedford and Robertson. “The Effectuof Various Factors on the Drained ‘Weight of Canned Red Cherries.’ Food Tech., 1955, Vol. IX, No. 7. p. 321. 121b1de. p. 321 71 AO‘M-‘ -.. Plants that maintain continuous and inplant inspection often have the processed cherries inspected twice. The first inspections are made immediately after processing. The product's probable grade is determined at this time on the basis of the score and the size of the cherries in the containers. 0n the hotpack, the net weight, vacuum, and drained weights are usually measured even though they do not affect the grade. It is on this grade that the processor must evaluate the quality level of his product in order to make immediate adjustments. The final inspection is often eliminated unless the product is being sold to the government or another buyer who demands certification of the grade. The final inspection is done by drawing representative samples out of a given lot in storage as contrasted to the probable in- spection that is done as the lot is being processed. Final inspection on the canned pack is made after the product has been in storage for at least ten days. The coldpack final inspections are often made the day after the cherries are processed. The curing period allowed between the probable and final inspections enable quality changes to take place in the processed product. Probable and final grades were taken from samples representing the same lots on both hot and cold packs in order to measure this change. Thirtybfivs hotpack samples of probable and final scores are used in this comparison. More samples would have been desirable but difficulty was met in obtaining lots that had both probable and final inspection made on them. 72 TIBLE XIII Change in Scores - Probable to Final - ...---” -—-v—w -..-“— m---~ —.—."CC—"----”fioou-"-~---~-“‘ovm-rovu - Character Color ...-fi .....- Jive. 561,2"Ee'f'r'1ne1 1 mug: Ave. asamfis.’ Final Change Score Score Score Score Hotmck 36.09 ' "77.115 23 ...- 1min “1.5.7.0 "'51? Coldmk 16. 1 11.0.2 _-.5. late ..9118. 1.}: FE] I Total Pack 16.50 37.12 1.? 17.10 18.21 15.5 "”- ' ' ”'fie’fgc'fi" """"“""" ' '*"_'fo't'a'1"s'é'6§e‘ -... Hotpack 16.26 36.66 1.1 89.78 92.20 2.7 Coldpaclt 36.71. 36.74 9_._9_ 3.512; 21.86 _._2 _ j Total Pack 36.50 16.70 .5 90.1») 92.03 1.8 A; .«m...-—.--...-—-.-.-...- o—-----.—-—.—.-—.—.--.-“~--.-m Quality change in the hotpack appeared to be much more noticeable. The hotpack total score increased 2.7% as compared to only a .9% increase in the total score of the coldpack. The average final total score on the hotpack is greater than that of the coldpack even though the coldpack had a greater average total score on the probable inspection. The greatest percentage change in scores for both hot and cold packs occurred in the color factor. The blending that occurs during the storage period is responsible for this improvement. Defects appeared to change very little from the probable to final inspection periods. The coldpack might show greater improvement if it were aIIOwod to set as long as the hotpack before the final inspection. CHAPTER VI SAMPLING AND STATISTICAL QUALITY CONTROL Raw and processed cherries are accepted or rejected and bought or sold according to quality on the basis of sampling inspections. Growers, processors, wholesale buyers and others who are involved in transactions of the particular products are all directly concerned with the accuracy with which a particular sample reflects the actual quality of a given load or lot. The first part of this chapter evaluates sampling plans and indicates how the reliability of a sample may be increased. The latter part of this chapter shows how quality control charts may be used in a typical cherry processing plant. Although the application I of these control charts are in the cherry processing industry, they are likewise applicable to other processing industries. Quality control charts are tools for management to use in.making economic decisions con- cerning resource allocation and utilisation. Those factors reducing or improving quality often may be easily detected by management through the aid of this tool and substantial savings may occur because of their dis- covery and adjustment. glaluation of Raw Product Sampligg The acceptability of a load is determined by the percentage of No. 1 grade cherries (Michigan revised Standards) that is found in a sample. A 500 gram or 1 pound sample is usually taken from a load by the federal- state inspectors regardless of the size of the load. 'If 88 per cent or more of the sample is No. 1 grade the load can be accepted for processing. When the quality of the raw product, as determined by the sample, is 73 7b questionable as to its acceptance for processing, the inspector might take another sample to verify the quality of the load. The past reputation of the grower may cause the inspector to be more or less cautious in gathering a random sample. Whenever a load's quality is based on a sample of that load, there exists a probability that this sample will not reflect the actual quality of the load. This sampling error can be reduced by increasing the size of the sample until the entire load is included in the sample. In acceptance inspection this sample is drawn randomly from the load and permits the measurement of the sampling error for different sample sizes. The probability of accepting a load of red cherries that contain a cer- tain percentage No. 2 or unclassified grade cherries for samples of different sizes is given in figure 7. The relationship between the per- cent defectives in the lots being submitted for inspection and the prob- ability of acceptance is called the operating characteristic of that particular sampling plan. The operating characteristic curves for raw product sampling indicates the probability of accepting or rejecting loads on the basis of the sample taken, if the actual quality of the loads are at different levels. The larger the size of the sample, the greater the ability of the sample plan to discriminate between lots of different qualities. In figure 7 a 16 per cent defective lot has a 1h per cent chance of acceptance if the size of the sample is 120 cherries and 12 per cent defectives or 14 defects are allowable for acceptance.1 If the size of this sample is 1120 cherries assumed equal to 1* or 500 grams 2&0 cherries assumed equal to 2* or 1000 grams 360 cherries assumed equal to Q4 or 1500 grams CO Probabi lity of Acceptan 75 Accuracy of Different Sampling Plans in Determining Raw Product Acceptability FIGURE 7 1.00 "0' n= 360,! \or{c = [4,2 .90. \. 080‘ ‘0 .70. H g 0604 I: Q 0501 E . O E .404 \ .304 ;\ .-\ n = 120 .20. ’.\— c = In .IOJ-H—O- '. . - n=21+0 ’. i c = 28 O 4 #4 L_ A a . e .}_O 0—. .__.-_ - 2 lb ,6 8 10 12 116 16 18 20 22 21+ 26 X Per Cent No. 2 - Michigan Revised ’1 =— Slzc. eé I‘M?”- C ‘7 Animal/Van‘s»! 0598‘“)? in San-r" 76 tripled (sample size ‘360 - allowable defects 1+2) then this same lot has only a 2 per cent probability of acceptance. The increased accuracy resulting from the doubling or tripling of the sample size is very important when payment is based on quality or a load is questionable as to its quality for acceptance. When time per- mits the inspector should increase the size of the sample taken from a grower's load. This is especially important when the loads are large and. unsatisfactory acceptance costs and quality payments are detrimental to the grower and the processor. As the sample size is increased the accuracy of the sampling plans increase at a decreasing rate. If the quality of the load contains 16 per cent No. 2 grade cherries then the probability of acceptance with different sample sizes are: Sample size II Chance of acceptance when shouldn't.2 120 I lb per cent 200 ”$5 per cent 360 I: 2 per cent Probability is used in reasoning from a known population to a ran- dom sample while confidence is used in reasoning from an observed sample to 1 ts unknown population. Sampling plans may be evaluated by determining the interval estimates of the quality of the load for different levels 0f confidence. In figure 8 the confidence interval for different sample sizes and for several confidence coefficients are determined. The con- fidence coefficients define the confidence that one might have that the quality of the load will fall between the confidence limits for that parthauler sample. If one makes the interval wider, he can have greater ‘ \o—o— 2 a" discrimination ability of the samples increased 9 per cent from gable sizes 120 to 2&0 and only 3 psr cent from sample 018” 240 ‘0 courlosuce RANGES FOR MW PRODUCT smpuuo "-360 N .240 NOIZO N7 c A. a h ‘s a h c. e s s h e s s h s b s h II 95 A" CONFIDENCE RANGES -- 99s ..- sax -— son - 80% "- 7070 mm}. I! 'ui‘x.‘ 78 confidence that his statement will be correct. An estimate that the quality of the load is between 97 and 83 score when the quality of the sample is 90 and the size of the sample is 120 is made with a .99 confi- dence coefficient. as can be confident that 99 times out of 100 that the quality of the load will be between these scores. If a lower con— fidence coefficient is chosen for the above sample then then interval will be much narrower. The ranges in figure 8 are only estimates of the '--_1 confidence range. The true confidence range must be computed by the use of the binomial distribution. As the scores move away from the 50 score the accuracy of the confidence ranges in figure 8 become more inaccurate. fl For this reason the confidence ranges of scores from 0—5 and from 95-100 are omitted because of the increase in the error at the extremities. The influence of sample size can be noted on width of the confidence intervals for similar levels of confidence by comparing A. B. and C in figure 8. when the sample is 90 score with a desired confidence coefficient of .90. the interval range decreases from 9.012 to 6.372 when the sample size is doubled and from 9.012 to 5.20% when the sample size is tripled. This again illustrates the importance of increasing the sample size to reduce sampling error. Evaluation_of Processed Product Sampling The processed product is inspected immediately after it is packed. The procedure usually used is to take a certain number of cans from each line at certain time intervals. The cans are then taken to the quality control laboratory, where the vacuum, drained weights, and net weights are checked and the contents are scored and graded. On the small sized cans the most common sampling procedure consists of checking 79 two cans every half hour. The No. 10 cans are usually checked by removing and testing one can every half hour. The 30 pound coldpack tins are scored and graded by removing a one pound sample from a tin every half hour. The processed product is scored and graded on a count basis as compared to a weight basis on the raw product. The sample on which a lot is graded is removed from what may be , considered as an infinitely large population. ‘we may assume, for fifififlflfia practical purposes that one No. 101 can is withdrawn and inspected out i of every thousand cans. The question may now be asked: if the lot has a certain per cent defective in one of the factors determining grade, 'j 9 what is the chance of accepting this lot as an A grade when it is a C Ew‘fl or D grade lot? These questions are answered in the operating character- istic curves in figure 9 which gives the probability of acceptance as A. C, or D grade when the actual quality of the particular quality character- istic in the lot is set at different levels. The grades of the processed products are determined by a point system in which any of the three factors of color, character, and defects can place the sample into a lower grade. If the quality is scored down more than 15 per cent on any of these factors, the product will fall into a grade C classification. If the quality is scored down more than 30 per cent on any of these factors, the product will fall into a grade Ilclassification. Because of this condition we may take each factor seperately and determine the probability of accepting it as a certain grade when the lot actual per cent factor defective is at different levels. If any of the factors fall into a lower grade, then the whole lot from which that sample is taken is considered to be in the lower grade. co Assume, for example, that the entire lot of cherries has been inspected and the average scores for the lot on color, character, and defects are 17, 36, and 30 respectively. Table XIV illustrates these scores and defines the grade into which the factors fall. ‘égvor 85 per cent of the color is good,.3%.or 90 per cent of the character is good and.&% or 75 per cent of the lot is free from defects. in can therefore say that the per cent defectives for color, character, and defects are 15, 10 and 25 respectively. Now the question arises as to what the chance is of accepting this lot as an A grade when in reality it is a C grade lot. We will assume that the size of the sample taken is 100 cherries. TABLE XIV Scoring System for Finished Product. .aA.Grade C Grade D Grade Color 20-—-2:2) l6—---lh 13---—-0 Character h0—{E§>34 33---28 27 ------ O Defects no..---3u 33.{:§}28 27------0 Total Score 100......85 W 69:-----0 In figure 9 color would have a 55 per cent chance of being accepted as an A grade and a #5 per cent chance of being accepted as a C grade. Character has a 95 per cent chance of being accepted as an‘A.grade and 5 per cent chance of being accepted as a C grade. Defects have a 2.5 per cent chance of being accepted as an A grade, 82.5 per cent chance of being accepted as a C grade and a 15 per cent chance of being accepted as a D grade. In the above lot, defects would be most likely to be the limiting factor in the quality determination of the lot. “MNMWY o snaps an --—-.-4n£r£cts)-—-->~----- / .. I ..... I I l I ........ w I u “......a. . ...m \m ... \ \ \ \ . on... ‘Q-eI--u*e-e'e'e-e'e'e- .. , . _ m . \ m _ . see fit? I.” i II I! | . s )- . n 4 mm A . . . a _ m ‘38; 038; new: 1 82 By increasing the size of the sample from 100 to 200 or 300 cherries, the probability of this sample predicting the actual grade of the lot is increased. This fact is well illustrated in figure 9. The operating characteristic curves for larger sized samples become steeper and show less dispersion at the extremes. The absolute size of the sample is much more important from the standpoint of quality determination than its relative size.1 The accuracy of the sampling plans increase at a decreasing rate as the sample size is increased. The selection of a quality acceptance plan depends on the risk and cost that the processor is willing to accept. This decision is usually based on the objectives of minimizing the sum of production costs, acceptance costs, and unsatisfactory product costs. The amount of inspection can be reduced, held constant or increased, depending upon the processor's estimate of the types and amounts of costs involved. The accuracy of any sampling plan cange described in the formation of an operating characteristic curve. This curve is therefore merely another tool which the processor can use in making economic decisions on sampling plans and procedures.“ Theory a£d_fipplication of Control Charts Control charts are tools that may be used by processors to detect unusual variations in processing operations. They provide management with guides for making economic decisions concerning the allocation, replacement, and adjustment of the resources or inputs going into ‘ ‘51‘ -..-m 3mm fact is well illustrated in Grants' book, Statistical leity Control pp. 316—317. '7' {for further reference in computing and understanding the formation of this Curve see 3. L. Grant, Stat. Qual. Control or A. V. Feigenbaum, anlity Contgg}. 83 xmocessing operations. This tool for guiding management may be defined as “a chronological graphical comparison of actual product quality characteristics with limits reflecting the ability to produce as shown by past experience on the product characteristics.'5 The inspections preformed on the processed cherry products provide the necessary data for the construction of control charts for different quality characteristics. The variation in quality of different character- istics in the past are used to predict what the variation should be in the future. If new technologies or practices, which reduce or increase quality variation, are adopted by a plant, then the expected variation in the future should also be reduced or increased. In the cherry process- ing industry defects, character, and color are quality factors which determine the grade of the finished product and are readily applicable to quality control charts. Other quality factors such as net weights, drained weights and vacuum are also measurable observations that can be easily analyzed graphically in control charts. Standard .devitidons are measurements of dispersion from the average of a number of observations. By the use of this measurement, predictions can be made as to where the observations of a given quality characteristic should fall. A large number of observations generally will form a normal distribution that is commonly described as a bell-shaped curve. If the pepulation is normally distributed, probabilities can'be determined as to the relative frequency that an observation will fall at given «tistances from the mean. Control limits are points on each side of the mean which discriminate between those observations that fall within a 5A} V. Feigenbaum, Qpalitz Control. Ic Grew-Hill, New York, 1951, 1st ed., ‘p. 121. ’ ". .. an _i 8b given distance and outside a given distance from the means. If the process is operating normally then the number of observations falling within these limits will approximate the probability that defines this range. If the process is not operating normally then this number of observations will not approximate the probability that is defined by the control limits. figure 10 more adequately describes the theory behind control charts. If a large number of observations are drawn from a population, the curve represents the distribution that should be obtained. Standard diviations of l, 2, and 3 on each side of the mean represent 68, 95, and 99.7 per cent of the area under the curve and defines the probability of a observation falling within this area. Three standard diviations are most commonly used to define the contol limits on a control chart. If control limits are set arthree standard diviations on each side of the mean, then three observations out of a thousand should fall outside these limits when the operation is normal. Control Chart - Based on 30" FIGURE 10 85 Observations do not necessarily have to fall outside the control limits before plant operations should be investigated. Trends in quality cflmracteristics or a concentration of these observations near the control limits may provide evidence of non-random elements occurring in the operations. Observations falling near or below the lower control limits cm above or near the upper control limits tell management when to investigate processing operations for possible causes of high or low defective quality character. If the quality character is extremely good as indicated on the control chart then the process should be investigated in order to find and adopt any element into the operation that may improve quality. If the quality characteristic is extremely poor as indicated by the control chart then the process should be investigated in order to find the cause of the poor quality and make the possible changes in plant operations that would correct this situation. The changes in quality should be examined critically until the factors causing these changes are discovered. The factors causing the poor quality should be eliminated and the factors causing quality improvement should be adapted whenever it is possible and economically feasible. The control limits of the control charts must be adjusted when any permanent change is made in the plant processing operations that will have an important bearing on the quality levels of the various products. Control charts may be maintained for every product quality character- istic that is measurable. The inspection service or plant quality con— trol personnel can provide the necessary information for the calculation m of these charts. Quality factors such as color, character, defects, and total score may be obtained from all processed product inspection samples. I 86 Color, character, defects and total scores are measured in standard units fbr all processed cherry products and therefore only four control charts, one for each quality factor may be used for the entire plant output. Separate control charts may be maintained of different groups of products or for each processing line in order to have more direct control over each of these quality factors. Drained weights, net weights and vacuum vary between the different types of products and cannot be aggregated together, therefore, each pro- duct must have a separate set of control charts for each of these factors. Additional work may be involved in keeping a separate set of control charts for each product or line'but this does help management to discover and find the cause of non-random quality variation by restricting the area of needed investigation. Some quality characteristics of the processed cherries are affected by weather and other seasonal conditions. In analyzing the records of several plants, seasonal trends were noted in the quality level of the different quality characteristics affecting the processed product. The weather factor is not controllable by man and therefore adjustments are needed in order to obtain control limits that reflect quality variation during any particular day of the season. In order to obtain reasonable control limits, trend lines must be obtained for seasonal changes and a standard error of estimate must be calculated from this trend line.- Control charts are presented on the following pages for the various qualitative characteristic that may be found in processed cherries. The control limits are computed as standard error of estimates on those factors that are affected by weather or seasonal conditions. The points plotted on the control charts for color, character, defects and total 8? score represent the points deducted from their possible scores. A score 0617 for color would be plotted as a 3 and a score of 88 for total score would be plotted as an 12. This reduces the size of the numbers needed in calculating control limits and provides the same amount of accuracy. The Observations plotted on the following control charts are taken from actual data of a processing plant during the 1955 season. In order to narrow the control limits on the plotted observations, the observations plotted are averages of four actual observations. The control limits for the individual observations can‘be obtained by multiplying the‘3a" for the grouped averages by the square root of the number in each group. In the case of these control charts, the control limits for actual observations may be obtained by multiplying the 30" control limits for the averages by ’Vnror 2. Three hundred sixty-four actual observations were grouped into 91 groups for the purpose of the following charts. Figure ll is a control chart for total scores for a processed product. The control limits are computed on the basis of a standard error of estimate calculated from the trend line which in this case indicates the quality becomes poorer as the season progresses. This standard error of estimate, as a measure of dispersion, is then used to determine the con- trol limits in figure 12 in which the trend is removed from these limits by the use of a moving average. This moving average can be computed as a certain number of days or as a certain number of observations. In figure 12 an average of 30 observations are used to obtain the adjust- ments. Ten observations are dropped and ten observations are added every time the adjustments are made. The moving average allows for seasonal changes and yet helps to keep these changes in check because of the residual effect of earlier quality levels. 88 Control Chart - Total Scores - Wi th Trend 6 FIGURE 11 204 . f Points Deduc ted \ O O ° to 50 3'0 zfo 50 sis 770 810 9?) Observations Over Time 6 In figure ll the trend line is 10.02 + .0371 and the standard error of estimate is 2.1109 and the control limits are set at 30" or ¢ 6.16 from the trend line. r-‘/_-—_— t — Poi nt's Deducted 89 Control Chart - Total Score - Moving Average FIGURE 12 , J 18' T ‘ 15“”le ‘ —-——-—— s s | s ' s ‘ ‘ 164 . u, s . . fl1-~--O--. --------- . ----- . ----slus-e-a, . ' as it. . . . . . . . O 0’ C O Q s ° 0 g _._n n..- 12' . ‘ " mad: . a" .. ‘. '. s u s 0 es s r, “...—u- -l— :m_ c . s ‘. 104 . 0 0‘ O O l .0 Bl 6w ‘“ “1 2‘ O fo {0 36 tio 50 6b 76 "We Observations over Time 90 Grade specifications can be incorporated into the control charts to provide information as to the probable grade that is being produced. These specifications may be represented as straight lines drawn across a control chart to distinguish between those observations that fall in E'grade classification and those that fall in the C or D grade classifi- cation. K 15 point deduction is allowable on total scores before the observation will fall into a C grade. This is indicated by line a - a in figure 12. Another grade line can be drawn at 30 to distinguish between C and D grades. Total score is composed of the scores of color, character, and defects. Therefore. the control chart for total scores provides management with an indication of when all three of these quality constituents should be investigated. Separate control charts for each of these three quality factors provide management with more direct guides in finding the cause of the non—random variations. Color is a quality factor that is affected by seasonal changes and therefore must be computed on a moving average control chart. In figure 13 the observations appear to be in control with little variation. A grade classification line can be drawn.through 3 on the vertical axis of this chart to distinguish this quality factor as falling in A grade or C grade. Such factors as improper soaking periods or improper cooking may cause color to fall butside the control limits. The influence of seasonal conditions on the color characteristic should be partially removed by the effect of control limit adjustments during the processing season. 91 Control Chart - Color - Moving Average 7 5 ., FIGURE 13 u w- . —— :3 . __ E 3 i- 2. - ”‘9‘”. 3;;ng.» \ O on” s s s s 3 b so. . ‘ so" u. 0—- s o s '3 2 0 . . . . s. ts. use so?“ .ss ...-9‘... fl.‘ —-~--— _ 1 u 0 L? ‘4. - _r A - HS 2 0 3’6 ho :35‘6 6!) 7'0 80 9o Observations Over Time Control Chart - Character - Moving Average8 FIGURE 1h 8 . we 0 ...: 8 6 . -- ————" U 0 s . , . , a “I .. .0 0.. .oo . e .... e o e ‘0 e o. s so. ‘3 {—4 f...‘.-‘: ...—ea .4.?_:....1. ...... .....—.— '4 9.... ‘ .0 .9 "D .00.. s O o. . s o be 2‘ o - a f0 20 30 E0 ?0 30'_'7—_T—'90 o ’0 Observations our Time 7 Control limits are 3 $1.20 3 Control limits are i s2.3o 92 Figure lb is a control chart for the quality factor "character.' In this chart grade discrimination lines can be drawn horizontally from 6 and 12 to distinguish between A grade and C grade, and between C grade and D grade. In figure 14 the variation of the observatipns are much greater than those for color. The last observation on the control chart is above the upper control limit and indicates that this process may not be in control. Investigation of the cause of poor character may reveal that the cherries are soaked too long or not long enough in the tanks before processing or that the processing equipment is not operating properly. PrOper length of soak is important in getting good character and better performance. Continued investigation may relate the cause back to the grower who may have delivered a load of cherries containing a high per cent of scald or other character defects. A character obser- vation falling below the lower control limit indicates that exceptionally good character is being maintained and investigations may reveal something about the process which can be adopted to improve the process in the future. ' Figure 15 is a control chart for defects. Grade lines may be drawn horizontally from 6 and 1? to indicate how the observations fall in relation to government specifications. In the above control chart one observation is extremely high and investigation may indicate that an excessive number of pits or pit fragments are present or that extremely . poor quality is being processed. The control chart may show that sorters should be placed between pitters or that more sorters should be put on the inspection belts. Good pitting performance or good raw product quality are two factors that may cause the observations to fall below the lower limits for defects. Mechanical difficulties and inadequate Ii 93 sorting labor are important quality factors. The speed of the processing operation can also be determined by the amount of defects passing over the belts as reflected in the score of the processed product. Control Chart - Defects - Moving Average FIGURE 15 12 ‘ } . “a.“ _.__________ o E 8 s ‘. ' Q 3 ° . s ’ ’ '. i o % t I . ' O ’9 6 d .. . . Inn-u:- Q . . . 1‘... . —. 0 O . . :—.— F O ' c «50 0 g .0 . ‘ ‘00 0‘ ° ' a h II—Q— _— —i-.—'...u-- . . . he 0. n O n . s i , s 2 1L . u. .0 .s' ’ “M o .-—— V *- .- .{. --— .— 10 20 30 no so 60 7o 83 9'? Observations Over Time V Ahi ‘- 9Control limits 'x' Z 5.03. L; J 94 The control charts for net weights, drained weights, and vacuum on lo. 303 canned R S P cherries are given in figures 16, 17, and 18. These quality factors do not have seasonal trend and therefore no moving adjustments are necessary and the control limits are computed as the standard diviation from the means. Net weights and drained weights may indicate that the fill weights may be reduced or increased or that some- thing is wrong with the filling operation. Drained weights are influenced by weather conditions from one year to another but do not have seasonal trends. vacuum in figure 18 exhibits a large amount of variation and probably could be brought under closer control by reducing the lag between the vacuum tank and the closing machine. 10 Control Chart — Net Weight (303 cans) FIGURE 16 20.001 18.00T 17.000. so . ‘00“ s «s .s .. ... ... . s m r—Oua’rksfis «3&4 --L—-a%~7‘fihg«ger -fl*041'3‘flh'*” s) {0 s ,‘, .aoce c “k e s s ‘5) .4 o ’ O 0 ’ ‘ o O 16.003 . 15.001 :‘ T 1 v‘ f —T j— 1— r-—r I 1 10 20 1O #0 50 60 7O 8O 90 95 ll ntrol Chart - Drained Weight (303 cans) Co FIGURE 17 13.00‘ “V ' . ' . O O .. . a 12.004 . ’ . . ...,s , o . so. so E l ' e ”s e O. . so. . .s , es s C; 11.00.” ...; ' l . . . . c'V'b—Tv-O s 0 g . . Os . . .0 s s O o O ‘ 10.00“ . O ‘ 0 J ' ' 9.00 . hrs—”W 1170““5'0' "W'flo‘fio‘flo Observations Over Time 12 Control Chart - vacuum (30? cans) FIGURE 18 1h.0 _, ——- r‘_'“ ‘L“h-$==""”T_'~ "* s 12.0 t e . O s s . . C 10.0 , so ' T s as . . O O .. O. 9.0 ‘ ' . f . . so .3 ...L..L. ..L °. ' ' 0 i ' ' a: — 0 .sO--e-w9__‘_—_—-—-— g 6 O o . O ' ‘ ’0' s . p-c T c . O s ' o O O “.01 . .. . .‘ ... O. in. - O 2.0 4 O o O ‘, 0+ ' f0 ""50" i0 E0 36 20 50 £30 00 Observations Over Time 11Control limits for drained weight 11.25 ¢Z.OO lZControl limits for vacuum 7. 50 ‘6. 50 96 The control limits used for any particular product must be cal— culated from emprical data of past operations. Those quality factors such as color, character, defects and total score which may exhibit seasonal trend should have their control limits based on the standard error of estimate while those which do not exhibit seasonal trend may have their control limits based on the standard deviation from the mean. It is a good practice to use the entire seasons output of one product to determine the control limits to be used for that product the following year. The control limits for products in the third year may be cal-a culated from the second years emperical data or from the average of the first and the second years data. It is necessary to use the previous year's observations in the calculation of the control limits because of the improvements that might have occurred in that year's operations as compared to the first year's operation. By weighing in the effect of previous standard deviations, the control charts from one year to another may become more standardized and very less from one year to. another. The standard deviations for the same quality factors vary considerably for different plants and products. It is, therefore, necessary to have the standard deviations for that particular quality factor and product calculated from several processing seasons before the desired standard deviations may be found on which to base the control limits. Operating characteristic curves are not only adaptable to sampling inspection, but they are also helpful in analyzing the probability of an observation falling within the control chart if the lot from which the observation is taken has a given percent defective. The computation 97 for obtaining the operating characteristic curve of a control chart that has both a lower and an upper control limit is well described by Freeman, Friedmal,‘mosteller, and‘Wallis.1'3 The poisson distribution tables by Molina are very helpful in the computation work.” I Figure 19 is an example of the results that one might get in calcup- lating an operating characteristic curve for a control chart that has upper and lower control limits. The chart is read by saying that if the actual lot contains a certain per cent defectives, what is the chance that the sample of cherries drawn from this lot will fall within the control limits. In the following example the control limits for a sample Of 100 cherries 19 86% at 5 and 18 defectives. This means that if the number of defective cherries is between 5 and 18 then this quality factor is in control. For a sample of 200 and 100 cherries the control limits are set at 10 and 36 defectives and 15 and 54 defectives. An increase in the sample size again results in increasing the discriminating ability of the sample. ‘ The statistical quality control charts discussed in this chapter are somewhat unique to the general concept of control charts. The manu— facturing industry usually has a much greater degree of control over the quality of its products and therefore can set a standard for its quality control chart which need only be changed when plant technology or quality specifications dnaage. The food processing industry usually has a lesser degree of control over the quality of its product because of the seasonal or weather conditions that influence quality before being received at A.“ 13Freeman, Friedman, Mosteller, and Wallis, Sampling Inspection, McGraw Hill Book Co., New Yerkzand London, lstyEd., 1949, pp. 63-66. I“Molina, E. c.,P1.30hv3 Exponential Binomial Limn, Table II, Van Nostrand 00., no., ew ork, 1942. 1.00 " .8g .60. .uq, Probability of Falling Within Control Operating Characteristic Curve for a Control Chart (with Upper and Lower Control Limits) FIGURE 19 9 a C. ' :‘mrl‘ file . 1 .. 100 .‘\§. ..... 3'9‘) \1 u" 300 lo 1% la 15* lé 20 52 Actual Percent Defective 98 28 so 32 99 the plant. The control charts in this chapter are illustrated with the hope that more food processors will see that the quality control charts used by manufacturing industries are applicable to the food processing industries by making adjustments by the use of moving averages to account for seasonal changes in quality due to weather and other conditions over which they have'no control. CHAPTER VI I SUMMARY AND CFNCLUSIONS (l) Considerable differences exist between processors in relation to their attitudes toward quality control and maintenance programs. Some processors provided the growers with more incentives and aid for pro- ducing quality cherries than others. Several plant managers stated that the quality that went into their plants was the quality that came out. In many plants the quantity of output was emphasized rather than the quality of output. (2) Processors have attempted to improve the quality of the raw product delivered to their plants by conducting grower meetings. using quality payment programs. employing fieldmen and promoting and engaging in other activities that may help the grower. Smaller plants were quite limited in their educational activities. Many different grower payment plans existed in the red cherry processing industry. The type of pay- ment plans used by a plant was closely related to locational groupings. Payment plans, or educational programs or services, appeared to have - very little influence on the quality of the raw product delivered to the plants. (3) The size of the operation of the individual growers seemed to have an effect on the quality delivered to the plant. Growers who delivered from 100 to 1000 lugs to a plant usually had higher quality than smaller or larger growers. The average size grower is probably better able to supervise harvesting and production operations than large growers. (h) An analysis of the growers' inspection records at three 100 101 Imocessing plants revealed that growers who delivered high quelity reduced their defects by reducing those that are controllable. Only 21.31 percent of the total defects shown in lots delivered by better quality growers were controllable as compared with 32.15 percent for the lower quality growers. This seems to indicate thAt the better quality growers had improved their quality by reducing those defects over which they had some control. (5) Raw product quality varied considerably from one day to another. The average daily quality change at a northern plant was larger than those daily quality changes in other locations. No consistent quality trends were noticeable at any plant. The locational factor has a higher relationship to a plant's average raw product grade than did the average size of the grower or the type of educational and payment plans. (6) The most important defects affecting the raw product quality were windwhip, scars and color. There was a closer correlation in the relative importance of these factors between a northern and southern plant than between two southern plants. Processors listed windwhip, scars, lugscald and color as the more important factors affecting the quality of the processed product. (7) Attempts at in-plant quality improvement varied from one plant to another. The average processing plant had two and three-fourths sorters for every thousand pounds of raw product going over the sorting belts in an hour. An one plant, only 1.3 sorters were used on the belts for every thousand pounds of cherries, while at another plant 6.67 sorters 'were maintained for every thousand pounds. Attitudes of management toward quality improvement and the quality of the raw product passing over the lines were responsible for this difference between plants. I 102 iflmn raw product quality was poor, the volume passing over the belts was‘reduced. (8) Sorters on belts varied in the amount of cherries they removed from the belts. It was not unusual to find one sorter removing three times as many cherries as other sorters in the same plant. The amount of pick-out varied between 2 and 9.6 percent of the volume passing over belts in 26 different plants. The average sorter picked out about 19.25 cmunds of cherries in an hour. The average percentage pick-out for the 76 plants was h.6¢. (9) Receiving, storing, size grading, pitting, weighing and other processing operations were important in quality improvement. Considerable differences existed between plants in receiving methods, handling practices and soaking periods. fieveral plents used two or more elimina-- tors in their processing lines. Many plants overloaded their pitters during the peak of the season. gome plants weighed every 500th small can while others weighed every 5,000th small can. The check for quality of the contents also varied extensively between plants. (10) 9everal plants mnintained continuous government inspection service on their processed products and other plants had their own specialized quality control personnel. The usual procedure in checking the product quality consisted of removing cans from the lines at certain time intervals. There appeared to be some correlation between the raw and finished product grades. The grade of the raw product appeared to he a poor index of the finished product grade because of the different quality characteristics used in grading the two products. There was an increase in the quality of the processed product between the time the probable and final inspections were made. This improvement was the L. 103 result of improved character and the blending of the color in the contents. (11) The need for quality improvement and quality control in the food processing industry has increased with the growth of competition. Increased cherry production has caused the cherry processors to enlarge their packs. This has resulted in more competition in pack sales among the packers. Government has bought cherries from packers to reduce the quantity-price squeeze, but the remainder of the pack must be put on a market where both price and quality will affect the sales. Quality maintenance may assure a processor of having a good outlet for this product. Several plants in the cherry processing industry have budlt such reputations. (12) The use of quality control techniques enables the processor to know what qualities are coming through his plant. Signs of possible plant improvements may be detected and adopted into the plant's regular operations. Difficulties in the process may also be detected and corrected. Economic decisions concerning the allocation and disposition of in-plant resources may be aided by the use of such tools. Quality control techniques are receiving increased interest by the food processing industry, yet up to this time few cherry processors in Michigan have used these tools. (13) More intensive study should be made of the actual quality improvement being made by a number of plants. The study that has Just been conducted was too general to make detailed in—plant and between- plant comparisons in quality and efficiency operations. Many differences exist between plants and additional study in this field is undoubtedly needed. (14) Information gathered in this study indicates that there may 10“ be considerable differences between inspectors in the grading of the raw product. Wide differences existed in the quality of the cherries graded by different inspectors at the same receiving stations. Raw product inspection records are available which could be used to further evaluete the importance of these differences. (15) Grading procedures for processed cherries presents a problem in statistical sampling. The grade of a lot is determined by multi- variate factors. A model should be developed to determine the probability of acceptance or rejection of s lot on the basis of these factors. This should prove to be a very intriguing problem for a statistics major. 105 BIBLIOGRAPHY Abbott, Laurence, Quality and Competition, Columbia University Press, New York, 1955. Anonymous , Aggi cultural Marke ti ng S erv.ig_e__s_tandard s - inspection Marketing Practices, United States Department of Agriculture, Public Lau'153:'83rd Congress, Washington, D.C., EffeCtive Date July 1, 195“. Anonymous, 'Are U; S} Canned Food Grades Commercially Suitable?‘ Egg food Packer, December, l9#3. 0—0-“- iation, Washington, D.C., June 1956. Anonymous, Canning‘Trade Almanac, Published Yearly by the Canning Trade Journal, Baltimore 2, Maryland. Anonymous, Michigan Cherry Surggy, Michigan Cooperative Crop Reporting Service, Lansing, Michigan, 19h9 and 1955. Anonymous, Questions and Answers on Government Ingpggtion of Processed Pruits‘ggd Vegetables, Misc. Publication No. 598, United States Department of Agriculture, Production and Marketing Administration, washington D.C., April l9h6. Anonymous, Standards for_ged Sour Cherries for Manufacture, United States Department of Agriculture, Production and Marketing Administration, Washington, D.C., Effective April 20, 19u1. Anonymous, U.S. Standards fgr_Grades_pf_Canned Red Sour (tart) Pitted Cherrieg, CFR Sec. 52:2h1, United States Department of Agriculture, Wishington, D.C., Effective June 18, 1949. Anonymous, U.S._§tandards for Grades of Fggggg Red Sour (tart) Pittgd Cherries, CFR Sec. 52:2h2, United States Department of Agriculture, Washington D.C., Effective June 18, 19h9. Bedford and Rdbertson, “The Effect of various Factors on the Drained Weight of Canned led Cherries," Food Technology, VOl. IX No. 7, 1955. Boger, L. L., Michigan Red Cherry Prices, Qpecial Bulletin 171, Michigan Agricultural Experiment Station, East Lansing, Michigan. 1951. Brice, B. A., The Measurement and Specifigation of 0010;, Eastern Utilization Research Branch, United States Department of Agriculture, washington, D.C. Feigenbaum, A.V., Quality»Contrgl, MCGraw Hill Book Co., Inc., New York 1951. Freeman, Friedman, Mosteller, and wallis, Sampling_Ingpection, MbGraw Hill Book Co., Inc., New York, let em, 191+9. Isl 106 Gaston and Levin., l'T‘ransporting Red Cherries in Water from Orchard to Processing Plant,“ gaggtggly Bulletin, Michigan Agricultural Experi- ment Station, Vol. 37, No. Q, East Lansing, Michigan. Gould, '2 A., ”Quality Emphasized at State Meeting,“ Food Packers, March,1955. Grant, 3. L., Statistical Quality Control, McGraw Hill Book Co., Inc., New York, 1952. Hills, C. H. and Whittenberger, R. T., 'Studies on the Processing of Red Cherries," Eyed Technology, Vol. VII, No. l, 1953. Knight, F. H., Risk Uncertainty and Profit, Houghton Mifflin Co., New York. 1 921 e Marshall, R. 3., Cherries and Cherrngroducts, Interscience Publishers, Inc., New York, 195“. Molina, E. C., Poisson's Exponential Binomial Limit, van Nostrand, Inc. New York, 19h2. Peterson, G. M. and Carleton, M. M., “Possibilities of Using Colored Lights for Detection of Cull Cherries," QuarterlygBulletig, Michigan Agricultural Experiment Station, Vol. 34, No. h, East Lansing, Michigan, November, 1951. Thompson and Spengler, Some Observations on the _Re1ationship of Quality of Fresh Sour Cherries to Their43rocessed Products and Effects of Processing of Various Types of Defects, United States Department of Agriculture, Production and Marketing Administration, washington, D. 0., June, 19159 APP HINDIX A QUESTIONNAIRE AND INSPECTION SHEETS USED FOR COLLECTION 0“ INF‘O’UMTION 107 '3 nrmmg cowrygmux. Department of Agricultural Economics Michigan State College The information asked for in this study is for research purposes only. Your name will not be made public and none of the information will be divulged in such a way that your answers can be individually recognized. (I) 'Name of business (2) Type of ownership ‘g- (3) Address _ , -.-- ..- ...... (1‘) Number of years in operation (5) Manager's name (6) Years at present Job THE NEXT FEW QUESTIONS PERTAIN T0 FINISHED PRODUCT AND NOT TO CHERRIES RECEIVED FROM THE FARMER (7) Throughout the last couple of years what factors have most often prevented you from putting up as good a pack as you would like to? Please list in order of importance. Item Hot Pack Cold Pack Excessive pits Undersize -————————- Lug scald -—-——~4——- Limb rub. -———————-— Color Defects variation in can weight Other: .— g g 0‘ (8) Have any of these affected the price you receive for cherries? Yes No -How much ___ i -7 (9) Do you try to put up a pack that exceeds the minimum requirement specified in government grade standards? Yes No . What score do you.attempt to obtain when packing "A" grade - I '0" grade ___ ? (10) What is your opinion on the following items with respect to present government inspection procedures? 108 -2- Item Hot Pack Cold Pack Are grade standards too high, not high enough, about right? Are samples taken often enough? Ye s_____ No___.~ Yes No“ Are samples taken large E [ enough to be accurate? Yes No Yes No Are there any important ‘ quality characteristics _ ' not now covered in the .J grade standards? If so ' “r‘ specigy. (11) What percent of your pack do you sell to customers who buy from you year after year? (12) Do you put up a different kind of pack early or late in the season in order to differentiate between the quality of cherries going into certain kinds of packs? If yes, explain THE NEXT FEW QUESTIONS PERTAIN T0 CHERRIE9 RECEIVED FROM THE FARMER (13) Do you pay a premium or bonus for high quality cherries? Explain ' 109 ~3- 0“) What minimum quality will you accept for processing? If this varies with use, variety or season please give details. (15) What is your opinion on the fOIIowing items with respect to present gpvernment inspection procedures? fire grade standards too high not high enough about right throughout the load to be accurate?‘ Yes No . If no explain: - .4 Are there any important quality characteristics not covered in the grade standards? Yes No _. If yes. specify PATRONAGE DmTA (16) How many growers do you buy from in a normal year . (17) What proportion of your farmers do not expect to consistently deliver high quality cherries? ,_,._”.,.% (18) 'What proportion of the cherries you receive are delivered directly to YOur plant by farmers , and what proportion comes from receiving stations __ , 7 (19) Is there a noticeable difference in the quality or condition of cherries received from each source? Explain - ‘_ (20) Ire cherries transported from receiving stations in water? (21) About how many hours pass from picking until these cherries are tanked? - ~'—-M (22) -h— lb you use ice in your tank? Yes No 110 (23) How many hours pass from tanking till cherries are finished processed? (24) Do you work with farmers in any of the following ways in order to get them to improve quality? Kind of Activity Fieldman Extent of activity, number of contacts or approximate cost w.-.-.- ' Growerguheetings flail contacts Other (25) (26) (27) , direct and what percent are made through Direct (28) I” You plan on any of the above in the future HOw much storage capacity do you have cold common ..“4 SALES DATA In a normal crop year (1953 for example) Hot Pack what percent of your sales are made brokers? Brokers;___ What percent of your direct sales are Mel! made in each of the following outlets in a normal crop year? (195?) a. Chain stores , b. Wholesale grocers c. Government (military, etc.) d. Institutional buyers (hotels, restaurants) e. Food processors (ice cream plants, bakers, preservers, pie mixers. etc.) L LLLL O ol‘.Pac* i Idrect Brokers -v--- (29) (30) (11) (32) V (13) (34) -5- To the best of your knowledge that percent of your broker sales go to each of the following in a normal crop year (1951) a. Chain stores b. Wholesale grocers c. Government (military, etc.) d. Institutional buyers (hotels, restaurants) e. Food processors (ice cream plants, bakers, Preservers, pie mixers, etc.) How many pounds of sour cherries did you handle in 1953_“_. -...n 1954 ? How many do you expect this year?__g ‘ Approximately what proportion of your pack graded “A" and what proportion graded 'C' in each of these years? was: % ....l__15 ......_, .....J W % * “A— Year Grade A mew-“ 1953 M‘ om“ firm...“ Q‘W‘. fiESected 1255---l What may be the cause for the variations MC or-..“ ”...... Amp—ru- . - --.——-»-——-———A 111 Cold Peg}; ofi‘ek s$sm L -u “ Grade C m ‘A.‘ - . i - (if there are any) above? I .‘ a.‘ Can you supply the following information on sour cherry sales for last year? ~- —— .....fflzsya___ - ._ o 4“” Mass. 9.1: fail: ioit‘eil."iaf1§.e‘7§f..a.n1.e.s_- -".‘~« COST DATA ~~m~w— Do you keep daily records on the volume of cherries processed, labor costs, and ingredient costs? Yes__~,Nq___. If yes, would you be willing to make these records available at some later date? 0*. 112 —6- PLANT INCPECTION Receiving: (35) Number of men employed in receiving? (16) Equipment used? ----w—c-“a o-c. ‘““ ..- ---~- .—-. __ m o--... -—————. n... N‘ - II (37) Describe method used in handling cherries up to point where they go onto belts; note particularly handling methods designed to 3 help prevent damage or maintain quality. .__-__.-_' u 111.0535: Fill out the following table: Line ‘fi- Length__. No. of P1cke£§‘”“fi6§5§1'EAEE’""’ Pick taken Number on line cherries pass per hour ____ over lineL;f[hr. ..,__ Pitting: (39) Number of pitters in operation u, ..-..s--- _- V, (40) Normal rate of use _ - -l_i.__- -7- (“1) Describe what happens to cherries after they leave the picking belt. (Complete a separate sheet for each kind of pack.) “--- - Filling weighing Exhausting Closing Cooking & codling Labeling & casing warehousing 92.11% Pitting‘ Filling Weighing Adding sugar etc. Covers Stamping Warehousing Description of Equipment including Capacigg 113 _Np. of Emplojees' involved per shift L 5 11h (UOZEO muwmppd 'llilll'll“|'l‘lll‘. 'ilIII'I'-.lll0‘ll'il‘ C C O C C O " "l'l'll"l|'""“| . C O I C C O . """!"'Illllll"' . O O O O I O O 'I'-‘II‘I‘I"I'I‘“" I O O n N I O O O O O C O . . O O O O O 0 . lllll'l‘l|"l""" O O O O O O 0 0 C O I O C C O C C O O O . O C . "l"'|l.l‘l'll|"l O O u o u o O O C H O O O I O C I 0 O O O 'iltigtlliiti{:!'! O O C u 0 O O “i'l‘ll‘i"i-“ii C 0 lol- u u o I I""!Il‘iiil.l§iii'l“' b u o u o ‘ll'!ii!‘sfli\‘§““ O I u o o o ll'l".|ll"1“|‘-' .3 o o o 0 AL 0 O C . !'lll'll"'q“i!"| O C I u 0 O O I‘..l‘l"'llq"""l . . . . NO 0 O O O ll“llll|llll!'l"'l' O O O O o C 0 'il‘1i|ln.lc'l!. puppets" . .. . O O n u . . II.‘I.IIIIIII'I‘IIII' 0 o o O O O O O u u o u 'IIIII'|'III'I.""-" I and ofiamcud so u so. u Imago M38... ('0 .m cm. .3 .HO «9 HQ¢M~U em 0.3 MO.“ QOflfiGHfiEflH wNHm 00% I -. wowemommmflo u I mom u opacfiuwfioo co munwao . I i. fifi? 3E mfiflfifl mmpmowsfi I * ..."I I I... I I“... I I w I I I". I I. h. I .. ... n......uIHIuInInIuHIuIquIHuImfimm Iul I Iul I IuI I I I". I IN I I I". I I W .HIHUIHIHIHIHIHIHIH m‘omfimlfimgom IuI I I“: I IuI I I .m. I I I“. I I .w I I .“I HIHIHIH m. I IBIHI m...“ mfirm Inflow u u u u n " «*wmIo Anvu u u u u n . u u “*MMIQNAUV « 3 “masonwpflmo .HHH IuIIruIIIuIIIWIIwIIwIIwBHngVwIIIIwIII IIII « u u u u “ "*wmIo AQVu " om .Q n u u u u “ "*MMImonV« o: u m6 .HH IulIlulIluIIIIwIIwII..“IIIMIfiamAMVIwIIIIWIMQVMIII u u u u u « 23an A5» u u n n u u " xflIfiSV“. .. ow .. n 580 .H IsIIIuIIIuIIIMIIImI IMIIhmefiAEw. II.I:”IIIIIIII IuIIInIIInIIIwIIHIIWIIIWIIImpmflomImmowmeImmmmwmm I”... I I“... ... I"; I ... ..e. ... I .m. I .. .m. .... ... .m. I I I mmfimflcmim mmwfifi Chum " u u . " ... u u I I Imam? mcmwemAWmmom a IHM IuI I I"... I I"! II a... I I. .w I. I. Im .... I290 .35 by . . m MmM moMpMGWWom. IuI I IuII Iul I I I”. I I I". I I I". I H I". I I I .I... ...I. HmMomsvapmmwvm mmmflmpm IuI IlullIulllIw IIImIImII IuIIII II IthanMvIeWszb. JII ..“I .........m .....Hm Hm MI...” muum II. Hun .....II memomswvufimwmm mg Hmhmm II I III III II“. I .. .m. I I .u. I I M: II I I II Ismswmswmmm MoI u u n u n u " Ewe 55.3380 IHII IullluIIIMIIMIIInIIIMIIIIIIIMmmHMymom MoIpmflMI « I u Ilnll IWIIMIIM IIIuI IIIII IIIpmwlnwuwmIMomepmr I I II I II I e383; 8:3 “mm 3% $38.33 I. I I I I I I I I 33 mmémmo BEBE Qfiav gen II mango I I I I I .oz .930 no.“ I I IF I I I LIEom :28 Emma $08 .02 .mmm I I I I I I I I I .07 .o .m :owpmfimnfisg mfipmfis: was $338an IIIIII I I I .02 $50 was 58 was no @353qu may”. .m. mags III! ‘0. l ... .I . u . I n . n O A s u o .I . a . ~ I I . to s a .e . I I - < c O . u I s O I n - o I o I 0 Q n I . a II. I I o 4 u .. . ... . . O . o . . . . i I I . . o . . . 0 . . n O I O I . . . . . . . . C .II .. ~ e n In. .. I O u :n' . O C J O I . a. .II I. . . . .a _ .e . o I n O ‘0 115 hopommwcH Hmfloflrfio I. l I’I III I III II ' IIIIIII III H 395me IIIIII I I I loom mpmowfippoo no mmpmnu sumac < mwmho mom I *x. coflpmflhmmflo SHE? eds mfififia 80.833 I * $3-0 AB" EBB *mmImmS: 3 mo HHH IIIII "I I IuI I In. I I I I .."I I Inn n nun H Um. H. mm... .... new. H mm H mm H H"mImImImImImImImIWm.Wm.wag II I I I...I I I"... I I." II I I I. JI I IuI "I .. ...u... I ..u.. I. I“... I I": I IHI I ..uquIquIqu...nIIIImQEHHIéHmom IIIII “IIJIIJIIIHHmIHmHHmIIuIInIInIIuIIuIIuwwwwwwgmwpgmfimm " IIIIIIIIIIIII I I I I I I I I I IuIEmééoI I “ IIIIJIIIIIJI "IIJIII " u " mSEE u n u u u u " *MMImNonu on a mo oHH " n u u u u I IuI I Iul I I I QUEEN I IuI ImpeHWmmHHI I IIIII "I I lul I Iul I I I I lul I Inl I I"! I lul I I"! I IuI « n u *MHIO 3v" u u n u u n u u. u u u u ” ISLEOW" om " H580 .H n u u "I "I I I"... I IuI I I“! Ilul I Inl I IuI I IuI DNHNWAM I... I I“ IIIIII I mmmmwme m m m m m m m m m m m m m menses Itoom M mmceoem ....u - .. I"- .. I"- .. .. I”- - I"- .. .."I I I“- - .. I“- I H.“ H .... H H H. ”HEW grime ewfimfim dam I l IuI I IHI I I“! I lnl I IuI I luI I .II...II... II... M.” H H...” ..IIIIuI I lul I IuI I lul I II I bmWomnvapmwMom Wow/HI " Honda mmmhmea "I I IHI I IuI I I"! I lul I lul I lul I I"! I I“! I lnl l lull lul I IuI I I I I moIflpImommNpflHomfilhm I n u n u u n u u u u u u u thé 553200 n IIIIIIIIIIIIIIIIIIIII I IIIIIIIIIIII I I I I I I I Mmmwm‘mom molpmflw. I » 98 «on? .....HWDmsm I I I I I I I I I I I I I IIIIIIII pmpommmfi Rama «ma mag. $3835 IIIII I I I I I I I I I I I I I I I In I 3.3 mflémm) SSE 35.3 ~58 BE fiwofi mumpsg .oz :28 pom I I IIIIIII I I I I I I IAH II. M H Hvrfiom $.30 58% 35% IIIIIIIII .ozdmm IIIIIIIIIIIIIIIIIIIII I .02 .9 .m 532935;?» mcfipgamz pcm :owpospomm 0,5333 .oz 3:80 $55323 mo .HHHIHfiHiHHH mamas :5sz (U9866 Form FbV—ll Revised 1-30-53 OFFICIAL INSPECTION REPORT UNITED STATES and MICHIGAN DEPTS. OF AGRICULTURE This report does not excuse failure to comply with any of the regulatory laws enforced by the State or Federal Departments indicated above. Insp. Point' Date Grower- Canner- Product° No. of Packages: Weight of sample - Percentage of defects: Number of worms in sample- Grade ° Remarks ' I, the undersigned, on the above date specified, made personal inspection of samples selected from the accessible portion of the load and do hereby certify that the conditions on said date, as shown by said sample was as stated herein. Inspector APPENDIX 'B CONTROL CHART DATA AND fiAM'PLTNG WALUATTON Number of Observation H O\OCD\JO\\J\¢\JNH NNNNNHHHHHHHI—‘H #uNI-doxomvmmtrwNI-o 25 {7 VP? W‘IJKwaU'~.-JKJKJUNNNN 117 Control Chart Data Points Deducted from; Color Defects Character Total Score 2.50 3.50 3.50 9.50 2.50 3.50 3.25 9-25 2.25 3.00 4.25 9.50 3.25 2.50 4.25 10.00 2.50 4.00 4.00 9.50 2.00 4.00 4.25 10.25 2.50 5.50 4.00, 12.00 3.00 4.75 4.00 10.75 2.75 3.50 1.50 9.75 3.00 3.50 4.00 10.50 3.00 4.50 4.00 11.50 3.00 3.75 4.00 10.75 3.00 3.00 4.00 10.00 3.00 3.50 3.50 10.00 3.75 7.25 4.00 14.00 3.00 5.00 4.25 12.25 3.00 3.50 4.25 10.75 3.00 4.75 4.00 11.75 3.00 3.25 4.00 10.25 3.00 5.50 4.00 12.50 3.00 7.25 4.25 14.50 3.00 6.75 4.00 ‘13.?5 3. 00 5.50 4. 00 12.50 3.00 4.50 4.00 11.50 3.00 3.50 4.00 10.50 3.00 3.25 4.00 10.25 3.00 2.00 4.00 9.75 3.00 2.00 3.75 8.75 3.00 3.25 3.00 9.25 2.25 2.50 3.00 7.75 3.00 3.00 3.00 9.00 3.00 6.00 3.00 12.00 2.25 6.75 3.00 12.00 2.75 2.75 4.25 9.75 3.00 2.50 4.00 9.50 3.00 4.00 4.00 11.00 3.00 3.25 4.00 10.25 3.75 6.25 5.75 15.75 3.00 5.25 4.50 12.75 3.00 4.00 3.50 10.50 3.00 4.50 3.75 11.25 3.00 4.75 3.25 11.00 3.00 4.25 3.50 11.25 2.50 2.50 4.00 9.00 3.00 2.00 3.50 7.50 Number of Observation 46 47 48 “9 50 51 52 53 54 55 56 57 58 59 60 61 62 m 64 65 66 67 68 69 7o 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 Color 2.25 3.00 2.50 2.50 2.75 2.75 3.00 2.00 2.00 2.25 3.00 3.00 3.00 3.00 3.00 3.00 2.25 2.25 2.00 2.00 2.00 3.00 2.50 2.00 2.00 2.25 2.00 2.25 2.25 2.50 3.00 2.25 2.00 2.00 3.00 2.75~ 2.25 2.75 2.50 2.75 3.00 2.50 2.00 2.25 2.00 Control Chart Data Points Deducted From; Defects 2.25 7.50 5.50 5.00 2.75 5075 6.75 7.00 5050 1' 0.50 9.50 11.75 8.50 8.00 7.25 7.75 5.00 5.00 4.25 6.00 6.25 5.75 6.25 7.75 4.50 6.00 5.00 5.75 5.25 8.75 7.75 8.00 5.50 5.25 6.75 5.75 10.75 7.25 7.75 60 50 6.00 8.50 4.25 5.25 7.25 Character 3.25 3.00 3.50 3.25 3.00 3.50 4.00 3.25 3.00 3.00 4.00 4.50 4.00 3.75 4.25 4.00 3.75 3.25 3.00 3.00 3.00 3.25 3.75 4.00 3.00 3.50 3.50 3.75 3.00 4.00 4.25 O O 88$ 0. OU‘OUtO OOOOO panwferp 7.75 13.50 11.50 9.25 8.50; 12.00 13.75 12.25 10.50 9. 75 16.50 19. 25 15.50 14.75 14. 50 14.75 11.00 10.50 ‘9. 25 11.00 11.25 12.00 . 12.50 14.00 10.00 11.75 10.50 11.75 10.50 15.25 15.00 13.25 10.50 11.00 15.25 12.50 17.75 13.25 14.25 13.25 13.00 14.50 9.25 12.00 15.25 118 Total Score I’— I 120 Control Chart Data for 303 Cans No. of Net Weight Drained weight vacuum Observation (ounces) (ounces) (pounds) 46 16.88 10.94 8.75 47 16.63 11.13 6.50 48 16.75 11.50 7.00 49 17.06 12.00 8.50 50 17.00 10.81 10.75 , 51 16.75 10.75 6.25 ' 52 16.38 11.56 6.00 53 16.94 12.50 6.75 54 16.88 11.50 6.00 55 16.88 11.75 4.25 56 16.75 12.19 4.75 57 17.00 12.81 4.50 58 16.88 12.06 3.75 59 16.75 12.50 2.50 60 17.00 13.25 4.75 61 16.81 12.00 4.00 62 16.88 12.19 4.00 63 17.25 11.00 8.00 64 17.19 11.63 8.75 65 17.13 12.44 7.00 66 16.63 11.88 5.50 67 16.69 12.44 6.00 68 16.50 11.81 8.00 ‘69 16.81 11.00 11.00 70 16.88 -11.94 7.00 71 16.69 11.19 9.00 72 16.88 11.06 9.00 73 16.31 10.88 11.50 74 16.81 11.38 8.50 75 16.69 11.75 7.25 76 17.13 11.25 10.00 77 16.88 11.88 8.75 78 17.13 11.81 9.25 79 16.94 12.06 8.25 80 17.06 11.25 12.00 81 16.75 11.50 8.25 82 16.38 11.44 10.00 83 16.88 11.50 10.50 84 16.88 10.88 10.00 85 16.69 12.19 13.50 86 16.06 11.50 9.50 87 16.44 11.38 6.25 88 16.81 10.88 11.50 89 16.88 12.31 10.25 90 16.88 10.25 12.50 121 Probabilities of Accepting Cherries for Processing with Different Sample Sizes and with Different Lot Qualities.‘ O .- M .— Sample Size 120 Sample Size 240 Sample Size 360 Percent Probability Percent Probability Percent Probability Defective of Defective of Defective . of Acceptance Acceptance Acceptance 02 1.000 02 1.000 02 1.000 04 1.000 04 1.000 04 1.000 06 .993 06 .999 06 1.000 08 .935 08 .977 08 .992 10 .772 10 .823 10 .860 12 .520 12 .490 12 .468 14 .298 14 .190 14 .132 16 .141 16 .050 16 .020 18 .060 18 .009 18 .000 20 .020 20 .001 20 .000 22 .010 22 .000 22 .000 24 .002 24 .000 24 .000 26 .001 26 .000 26 .000 28 .000 28 .000 28 .000 30 .000 30 .000 30 .000 I”Afllmm'tble percent defective is 12. CONFIDENCE'RANGE FOR RAW PRODUCT SAMPLING N I 120 122 level of confidence c-fi - ma--.- ‘ -n»—~—-—p m¢~--c—4- 80 90 95 99 $2.551 $3.274 at 3.900 a: 5.126 2. 779 3. 566 4. 249 5. 585 2.986 3.831 4.565 6.000 3.176 4.075 4.855 6.381 3.349 2.397 5.120 6.729 3.511 4.506 5.368 7.056 3.661 4.698 5.598 7.357 3.802 4.879 5.813 6.640 3.936 5.050 6.017 7.908 4.061 5.211 6.209 8.161 4.179 5.363 6.390 8.398 4.291 5.506 6.560 8.622 4.396 5.641 6.721 8.833 4.496 5.769 ‘6.874 9.034 9.591 5.891 7.019 9.225 4.681 6.006 7.156 - 9.405 4.767 6.116 7.287 9.578 4.849 6.221 7.413 9.742 4.925 6.320 7.530 9.897 4.999 6.414 7.642 ‘10. 5.068 6.503 7.748 10.183 5.133 6.587 7.848 10.314 5.196 6.667 7.944 10.441 5.255 6.743 8.034 10.559 5.310 6.814 8.118 10.670 5.363 6.881 8.199 10.775 5.582 7.162 8.534 11.216 5.733 7.356 8.765 11.520 5.822 7.470 ~8.900 11.698 5.851 7.508 8.945 11.757 1.... *“~ 123 Confidence Range for Row Product Sampling N = 240 “...-.... % level of confidence 9core of - ~ -*—-3- Sample 70 _. 80 9O 95 99 95 :1 1.458 :4 1.804 $2.315 $2.758 d: 3 .625 94 1,538 1,965 2.522 3.005 3 949 93 1.706 2.111 2.709 3. 228 4. 243 92 1.814 2.246 2.881 3.433 4. 512 91 1.913 2.368 3.038 3 620 4.758 90 2.007 2.483 3.186 3 796 4.989 89 2.092 2.589 3.322 3.958 5.202 88 2.173 2.688 3.450 4.110 5.402 87 2.249 2.783 3.571 4.255 5.592 86 2. 321 2.872 3.685 4.391 5.771 85 2. 388 2.955 3.792 4. 518 5.938 84 2452 3.034 3.893 4.639 6.097 83 2. 512 3.108 3.989 4. 752 6. 246 82 2.569 3.179 4.079 4. 861 6. 388 81 2.623 3.246 4.166 4.963 6. 523 80 2.674 3.310 4.247 5.060 6.650 79 2.724 3.371 4.325 5.153 6.773 78 2.770 3.429 4.399 5. 242 6.889 77 2.814 3.483 .469 5. 325 6.999 76 2.856 3.535 4.535 5. 404 7.102 75 2.896 3.584 4.598 5. 479 7.200 74 2.933 3.630 4.658 5. 549 7.293 73 2.969 3.674 4.714 5.617 7.383 72 3 003 3.716 4.768 5.681 7.466 71 3.034 3.755 4.818 5. 740 7.545 70 3.065 3.792 4.865 5.798 7.619 65 3 190 3.947 5.064 6. 034 7.931 60 3.276 4.054 5.201 6.198 8.146 55 3 326 4.117 5.282 ,6.293 8.272 50 3 343 4.137 5.309 6.325 8.313 Score of Sampl 95 94 9? 92 91 90 89 88 87 86 85 84 81 82 81 80 79 78 77 76 75 74 73 71 70 65 60 55 50 Confidence Range for Raw Product Sampling 124 % level of confidence 70 80 9O 95 99 :1: 1.191 $1.473 361.890 *2.252 2.960 1.297 1.605 2.059 2.453 3.225 1.393 1.724 2.212 2.636 3.464 1.482 1.834 2.353 2.803 3.684 1.562 1.934 7.481 2.956 3.885 1.639 2.027 2.602 3.099 4.074 1.708 2.114 2.712 3:232 4.248 1.774 2.195 2.817 3.356 4.411 1.837 2.273 2.916 3.474 4.566» 1.895 2.345 3.009 3.585 4.712 1.950 2.413 3.096 3.689 4.849 2.002 2.477 3.179 3.787 4.978 2.051 2.538 3.257 3.880 5.100 2.098 2.596 3.311 3.969 5.216 2.142 2.651 3.401 4.052 5.326 2.184 2.703 3.468 4.132 5.430 2.24 2.752 3.531 4.207 5.530 2.262 2.800 3.592 4.280 5.625 2.298 2.844 3.649 4.347 5.714 2.332 2.886 3.703 4.412 5.799 2.364 2.926 3.755 4.473 5.880 2.395 2.964 3.803 4.531 5.955 2.424 3.000 3.849 4.586 6.028 2.477 3.066 3.934 4.687 6.160 2.502 3.097 3.973 4.734 6.221 2.604 3.223 4.135 4.927 6.476 2.675 3.310 4.247 5.060 6.651 2.716 3.361 4.313 5.138 6.754 2.730 3.378 4.335 5.164 6.788 Finished Product Grading Probabilities For Different Sampling Plans 125 Samle size 100 Sample size 200_ ... Sample size 3.0L Percent Grade Grade Grade ._____ Defecjges A B C A B C A B C 00 1.000 .000 .000 1.000 .000 .000 1.000 .000 .000 00 1.000 .000 .000 1.000 .000 .000 1.000 .000 .000 04 1.000 .000 .000 1.000 .000 .000 1.000 .000 .000 06 .999 .001 .000 1.000 .000 .000 1.000 .000 .000 08 .992 .008 .000 .999 .001 .000 1.000 .000 .000 10 .951 .049 .000 .987 .013 .000 .996 .004 .000 12 .844 .156 .000 .904 .096 .000 .939 .061 .000 14 .669 .331 .000 .690 .310 000 .712 .288 .000 16 .467 .532 .001 .406 .594 .000 .367 .633 .000 20 .157 .830 .013 .062 .937 .001 .027 .973 .000 22 .077 .883 .040 .017 .974 .007 .004 .994 .002 24 .034 .870 .096 .004 .957 .039 .000 .983 .017 26 .014 .800 .186 .001 .878 .121 .000 .919 .081 28 .005 .685 .310 .000 .731 .269 .000 .764 .236 30 .002 .546 .452 .000 .534 .466 .000 .528 .472 - 32 .001 .405 .594 .000 .337 .663 .000 .291 .709 34 .000 .280 .720 .000 .182 .818 .000 .095 .905 36 .000 .181 .819 .000 .085 .915 .000 .030 .970 38 .000 .109 .891 .000 .034 .966 .000 .006 .994 40 .000 .062 .938 .000 .012 .988 .000 .001 .999 42 .000 .033 .967 .000 .004 .996 .000 .000 1.000 44 .000 .017 .983 .000 .001 .999 .000 .000 1.000 46 .000 .008 .992 .000 .000 1.000 .000 .000 1.000 48 .000 .004 .996 .000 .000 1.000 .000 .000 1.000 50 .000 .002 .998 .000 .000 1.000 .000 .000 1.000 KPPENDIX C PACK STRTISTICS 127 ooo .43. .oH oo .mfi own 63 ooo .oo: .o oo 82 ooo .E 32 ooo .93 .5 oo .NHN ooo .rofl ooo .92. .2 oo .oNN ooo .3 ooo” ooo .38 .8 oo .82 o2 .42 ooo .43 .2 oo 82 ooo .8 $3 ooo .on .1: oo 4.2 oNH .oS ooo .mp4 .6 oo .ofi oom .8 «mm: ooo .88 .HN oo .oS SN do.” ooo do: .2 oo .94 ooo so RS ooo .mom .om. oo .oS ooo .63 ooo .91. .NH oo .oS ooo .oo omS ooo .4? .om oo .9: ooo .ooH ooo .2o .3 oo .NS oom .oo SS ooo .o: .om oo .2: ooo .HS ooo .owo .2 oo .2: ooo .oo 33 ooo .mofi .3 oo .ofi o& .oo ooo .oom .o oo .8: oom .3 SE ooo .48 .8 oo .Hom ooo .63 ooo .omo .3 oo .oom oom .oo 32 ooo .ooo .3 oo .ooo ooo .mo ooo .8: .o oo .48 ooo .3 $2 ooo Jon .3 oo .82 8: .N: ooo .ooo .6 oo 82 ooo .om. 32 ooo .omm .8 oo 4.2 ooo .oo ooo .omo .H oo .8; ooo .2 SS ooo .83 .3 oo .ooH ooo .moa ooo .omo .4 oo .ooH oom .3 ~42 ooo .96 .8 on .8 oo: .5 ooo .9} .N oo .8 o2. .8 32 ooo .NS .o oo .3 ooo .ooH ooo .ooo .N oo .3 oom .3 33 ooo .8.“ .o om . no oi .oo ooo .o? .H oo .8 ooo .% on? ooo.omo.m oH.om omm.oo ooo.Hmm oo.oo ooo.NH mrofl .l I - -. :83 283 . nofiuofivoum mo ofiad> noe\eoHpm codpodvoum coapodvopm no 05Hm> coa\ooupm :owaofivopm you» @543 9852: . 5.onon -11'3'-‘ ‘1‘ l :anofiooum huuono com No. of Observation \OmQO‘tm‘F’ri-I' Control Chart Data. Net Weight (ounces) 16.81 16.63 17.06 17.13 16.88 16.9% 16.94 17.06 16.88 16.94 17.06 16.81 16.90 16.88 17.11 16.88 16.88 17.13 16.81 17.00 17.00 17.00 17.00 16.75 16.69 16.19 16.81 17.00 16.81 16.56 16.63 16.53 16.75 17.00 17.00 17.00 16.93 16.50 16.69 16.63 16.63 16.60 16.38 11.88 16.75 for 303 Can: Drained Weight (ounces) 11.19 10.81 11.19 11.19 11.03 10.56 11.63 11.13 10.88 11.00 11.19 11.00 11.25 12.00 11.56 11.19 11.00 11.00 10.81 11.13 11.13 10.n4 11.06 11.75 11.nu 10.13 10.06 10.50 10.38 9.88 10.81 11.81 10.50 10.81 11.13 11.50 10.88 11.40 11.56 11.75 10.94 11.56 11.00 10.08 10.38 vacuum (pounds) 3.25 5.25 3.00 5.50 3.35 6.00 3.75 7.50 7.00 6.50 6.25 2.75 5.25 7.25 9.75- 6.00 5.75 6.50 6.00 “.50 7.00 6.50 6050 3-25 5025 7.00 8.75 9.75 7.50 9.75 “.50 3.75 8.50 6.00 7.00 4.50 6.50 1.75 1.00 2.75 0.50 “.50 7.25 6.75 6.75 119 (Millions of Pounds) Production and Disposition of the Sour Cherry Crop % Brined % Frozen Tbtel Lbs Proc. % Canned Lbs Brined Lbs canned Lbs Frozen Year MBNOCDHQNOWHQJHMHOBNJQN d>v3m>oxd-bu4.4.4«4 ° GIOIGJFi .4 (\BOQMQOBQJHNBNL'DWO‘CDONF1 00MBIN-3VNO\O\ONOI\O\#I\H 0‘30 HH NNHMC‘NNNN NNmm N 4&xococuCACDracxxoxnra:rkuxnaaxn41c3c>00c: O O O O O O O O O O O O O O W)0\C>F:b~m)v\d>d)d)b~O\C>¢)F‘rib~Q)C>U\C> a)b~b~q2a>b-b~a>u>o~a>w><>unb~b~b~b»m>w>b- c1F+O\o\ozq>w>w1b~m>ozb-O\o\vxc>wbaJC>u>«> O O O O O O O O O 0 O O O O O O O O O O O a)()O\C>Q)0“b~V\F\r4(\F‘C>O\F\btg-3I\r4GD U\UNu>v\v>a)d>a>~co\ozr4oxo1r4a) r441