Stochastic analysis of TPS: expose and eliminate variability by highly specifying WCP

Abstract

A production system is designed with three elements: works, connections and pathways (WCP), which cover how people work, how people connect and how the production line is constructed, from the point of view of a flow unit. The design principle of highly specifying WCP leads to two operations pillars that support the Toyota production system (TPS): just-in-time (JIT) and autonomation. Some stochastic models are proposed to verify the logic of highly specifying work for JIT. The stochastic models show that works are highly specified to reduce the variability whereas highly specifying connections and pathways without resource pooling expose variability in works as low inventory exposes the hidden problems in a factory. To achieve JIT, highly specifying works in each workstation is fundamental; the key solution is to reduce the variability of works. The information entropy theory is used to verify the logic of highly specifying WCP for autonomation. Highly specifying WCP can reduce the information needed by workers to identify and solve problems in the flow unit.

Keywords:

• Toyota
• Resource pooling
• Just in time
• Autonomation
• Stochastic models
• Information entropy

Acknowledgments

The authors gratefully acknowledge the support of the NSFC (Grant 70501029). We thank the reviewers very much for their valued suggestions.