An integrated Delphi-MCDM-Bayesian Network framework for production system selection

Abstract

Several attempts are needed to choose the most compatible production system for achieving the desired manufacturing outputs. The significant role of manufacturing strategy deployment is selecting the production system best suited for a manufacturing firm. The appropriately chosen production system (strategic process choice) facilitates a firm to produce “order winning” outputs and provides a production competence to achieve business success. This research presents a novel framework to determine the compatible production system for a manufacturing firm. An integrated three-stage Delphi-MCDM-Bayesian Network (BN) framework has been proposed. The process choice criteria (PCC) considered for deciding production systems are identified through an in-depth literature review and then validated by experts through a Delphi method in the first stage. It resulted in the determination of twenty-six PCC. In the second stage, the multi-criteria decision-making (MCDM) based voting analytical hierarchy process (VAHP) method is adopted to determine each criterion's relative importance for a firm. The relative weights obtained are then used as input for the machine learning (ML) technique- Bayesian network (BN) in the third stage. The BN model quantifies the selection probability of production systems. The proposed Delphi-MCDM-BN framework is demonstrated using a real-life case of a “hydraulic and pneumatic valve” manufacturing firm to select a suitable production system. The three-stage framework is a novel contribution to the literature, which can be used by researchers, practitioners, and manufacturing strategists to choose an appropriate production system for any manufacturing firm.