Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name

Computer Science


Theoretical Statistics and Mathematics Unit (TSMU-Kolkata)


Mukhopadhyay, Anis Chandra (TSMU-Kolkata; ISI)

Abstract (Summary of the Work)

Irrespective of the type of product, evaluation of conformity to specified requirements of its quality characteristics is an integral part of quality assurance. Although they form a set of necessary verification activities almost at all stages of production, these activities, known as inspection do not add value to the product on their own and are to be kept at their minimum. The sampling inspection where a portion of a collection of product units is inspected on a set of characteristics with a view to making decision about acceptance or otherwise becomes relevant in this context.The number of elements of the set of characteristics to be verified for this purpose is unlikely to be just one in most of the practical situations. We may call this as multiattribute inspection as employed for verification of materials procured from outside and further at all stages of production, through semi finished and finished or assembly stages to final despatch to the customers. At all such stages consecutive collections of products called lots, are submitted for acceptance or alternative disposition.In this context one observes that the defectives with respect to different characteristics may in some situations be considered as jointly independent, whereas in some situations, occurrences of one type of defect may preclude occurrence of any other type. In some other situations, it is more natural to count number of defects for each unit of product so that the quality of a lot is expressed as an ordered set of average number of defects per unit for each type of defect instead of proportion defectives in an aggregate.The present exercise attends to the designing of the procedure of sampling inspection by attribute with the above scope and purpose in mind. One may call them as multiattribute single sampling plans(MASSP).The first objective is to establish a sampling scheme in line with available international standards tabulated on the basis of Acceptable Quality Level(AQL) in such multiattribute situations. These Standards in general prescribe that separate plans are to be chosen for the different classes of attributes. We examine the consequences of constructing a sampling plan by this prescribed method in a multiattribute situation. We first consider the effective producer’s risks. Secondly, it has been thought as reasonable to expect that the OC function should be more sensitive to the changes in the defect level of more important attributes, particularly, in a situation, where unsatisfactory defect level occurs due to more serious type of defects. A measure of sensitivity has been defined, for this purpose. Plans constructed with all possible practically useful combinations of AQL for three attributes have been examined for both the as dicussed. It has been found that the features observed depict a picture far from the ideal on both the counts for most of the plans.Further, it has been discovered that these two desirable properties remain generally dissatisfied inherently absent for the class of plans (we call these as MASSP of C kind), where we accept a lot if the number of defectives observed in the sample is less than the respective acceptance numbers stipulated separately for each attribute depending upon the AQL.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Included in

Mathematics Commons