An integrated Bayesian–Markovian framework for ascertaining cost of executing quality improvement programs in manufacturing industry

Typically, the budgetary requirements for executing a supplier’s process quality improvement program are often done in unstructured ways in that quality improvement managers purely use their previous experiences and pertinent historical information. In this backdrop, the purpose of this paper is to ascertain the expected cost of carrying out suppliers’ process quality improvement programs that are driven by original equipment manufacturers (OEMs).

Using inputs from experts who had prior experience executing suppliers’ quality improvement programs and employing the Bayesian theory, transition probabilities to various quality levels from an initial quality level are ascertained. Thereafter, the Markov chain concept enables the authors to determine steady-state probabilities. These steady-state probabilities in conjunction with quality level cost coefficients yield the expected cost of quality improvement programs.

The novel method devised in this research is a key contribution of the work. Furthermore, various implications related to experts’ inputs, dynamics related to Markov chain, etc., are discussed. The method is illustrated using a real life of automotive industry in India.

The research contributes to the extant literature in that a new method of determining the expected cost of quality improvement is proposed. Furthermore, the method would be of value to OEMs and suppliers wherein the quality levels at a given time are the function of quality levels in preceding period(s).