Bayesian belief network for assessing impact of factors on army’s lean–agile replenishment system

Abstract A dynamic time-separated lean–agile spare part replenishment system can prove beneficial to the army by being efficient (cost saving) during peace and effective (assured availability) during war. The logistics echelons must have certain attributes in order to implement such a dynamic replenishment system. The purpose of this article is to identify the factors/attributes that are necessary in a spare part replenishment system of vehicles and weapon platforms in order to implement a time-separated lean–agile strategy through a systematic literature review. Furthermore, the article will investigate the impact of these factors/attributes, individually and collectively, on overall system performance. This will enable logistics managers to focus only on the factors that have greater impact on the system. A model explaining the effects of various contributory factors/attributes on the overall logistics system has been developed through a comprehensive literature review, experts’ judgments and inputs from practising logisticians in the military field. The article then models the system using a Bayesian belief network (BBN) on Netica software. After the development of the model using Netica, a sensitivity analysis based on the mutual information criterion is conducted to identify the critical factors that most significantly affect a dynamic lean–agile spare part replenishment system. The study addresses the identified need of applying BBN to model an uncertain and complex military logistics domain.