Graduate Thesis Or Dissertation

Determination of optimum raw product sampling procedures with special reference to green beans for processing

Public Deposited

Downloadable Content

Download PDF


Attribute NameValues
  • Most raw product received at processing plants in Oregon is sampled to determine quality characteristics as a basis of grower payment. Raw product sampling procedures must provide the processor with a means of sampling which will be sufficiently precise to satisfy himself as well, as his growers, and thus avoid misunderstandings relative to the true grade of product delivered. How well a given sampling scheme will accurately provide this information depends upon the amount and nature of variability among characteristics of the product within a given load and the size and type of sample being drawn, Through the use of probability sampling theory it is possible to judge the precision of a sampling procedure by examining the frequency distribution generated for the estimate if the procedure is applied repeatedly. Based upon the knowledge of the frequency distribution of the estimate, it is possible to determine the range within which, as an example, 95 per cent of all possible samples would fall. The processor not only is interested in the precision and accuracy of his sampling procedures, but he also is concerned with costs of sampling. This study considers costs of sampling as well as precision in determining optimum sampling procedures which will minimize cost in achieving a given level of precision or maximize precision for a given cost. Green beans for processing were used in the application of sampling theory. Green beans are delivered to the processing plants in tote bins loaded on trucks. Two-stage sampling procedures are used where a sample of totes is drawn from the truck and a sample of beans is drawn from each tote selected. Repeated sampling was used to estimate the variation existing among primary and secondary sampling units. Three sampling schemes were employed to estimate components of variation: In Sampling Scheme A two 1O pound secondary units were drawn from each of two primary units. Sample Scheme B consisted of drawing two 3O-pound secondary units from each of two primary units. Sample Scheme C consisted of drawing a single continuous sample drawn from each of two primary units. The study was designed so that a comparison might be made of (1) the precision achieved by various numbers of primary and secondary sample units, (2) the effect of the size of the secondary units on among primary and among secondary unit variation, (3) the completely random and the random systematic (continuous) method of obtaining the secondary unit. Costs involved in sampling raw product were estimated by economic-engineering techniques. Costs were divided into two components - - costs associated with the drawing of the primary unit, and costs incurred in drawing and grading the secondary unit. Grading costs were a major portion of total estimated costs of sampling. Variance and cost estimates were brought together for use in the final analysis as follows: (1) optimum sampling plans were determined for each of the three sampling schemes for selected levels of precision; (2) comparisons were made of the optimum for each of the three schemes to determine the scheme which provides selected levels of precision at the lowest costs; (3) generalizations were made with regard to other sampling schemes involving sub-samples of sizes other than those included in the study and (4) a comparison was made of the simple random and the continuous method of selecting the sub-sample. Based on the range of precision and confidence considered in the study, sampling Scheme A appears to provide a least cost method of achieving given levels of precision and confidence.
Resource Type
Date Available
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Academic Affiliation
Non-Academic Affiliation
Rights Statement
Peer Reviewed



This work has no parents.

In Collection: