Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name



SQC and OR Unit (Kolkata)


Mandal, Pathik (SQCOR-Kolkata; ISI)

Abstract (Summary of the Work)

In the field of Quality Control (QC), the term 'control' is used to mean many different things. For example, it may refer to an isolated control action (e.g. a process adjustment), a particular type of control architecture (e.g. feedback vs. feed forward control) or even a control discipline (e.g. Deming's PDCA cycle). So will be the case here. But the exact meaning will be clear from the context.Since the development of control chart by Shewhart [1] and the method of acceptance sampling by Dodge and Romig [2] during 1920's - 1940's, the field of quality control has been greatly enriched by the development of many new principles and methods (in particular, the statistical methods for QC). Notable among these are the following:(i) Development of the new philosophy of Japanese management during 1940 - 1950 [3];(ii) Development of the concept of Total Quality Control (TQC) by Feigenbaum [4] in late 1940;(iii) Widespread use of the simple tools of QC from 1950 onwards and the pioneering efforts of Deming [5], Juran [6] and Ishikawa [7] in achieving the same;(iv) Development of the concept of robust design and the statistical methods for achieving the same by Taguchi (see ref. [8]) during 1970's and(v) Recent popularization of the Six Sigma breakthrough strategy for process improvement by Harry (see ref. [9]).Of course, the above efforts have been greatly supported by many researchers and practitioners from academia and industry. In what follows, there will be many opportunities to have a look at some of these contributions, particularly in the area of Statistical Process Control (SPC).1.1. Classification of control activitiesThe QC tasks performed at various stages of product/service realization before shipping can be broadly classified as shown in Figure 1.1. It should however be noted that all the six activities are not of equal importance. There is an inherent hierarchy among them. Anticipating customers' usage conditions of a product and making it robust so that it performs well under all such conditions are obviously more important activities than the activity of inspection for product acceptance/rejection. It is now widely recognized that the former has got much higher potential to influence customer satisfaction than the latter (which, actually, is aimed at minimizing customer dissatisfaction). In fact, one can say that the control objective of each of these three stages of product realization is to reduce or even eliminate the need for control at the downstream.This work is concerned essentially with the two control activities at the middle of Figure 1.1, i.e. shift control and drift control. These two terms are not unknown to the QC community at large. But here they will be used in a slightly different fashion as explained in the next section.1.2. Terminology1.2.1. Shift and driftThe core of SPC is control charting for detecting process abnormalities or 'shifts'. The presence of shifts refers to any type of process behaviour, which is different from the expected behaviour under an in-control state of the process. It is to be noted that ‘shift’ does not represent any particular type of variation.


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