A generalization of the Theory of Constraints: Choosing the optimal improvement option with consideration of variability and costs

Abstract

The Theory of Constraints (TOC) was proposed in the mid-1980s and has significantly impacted productivity improvement in manufacturing systems. Although it is intuitive and easy to understand, its conclusions are mainly derived from deterministic settings or based on mean values. This article generalizes the concept of TOC to stochastic settings through the performance analysis of queueing systems and simulation studies. We show that, in stochastic settings, the conventional TOC may not be optimal, and a throughput bottleneck should be considered in certain types of machines at the planning stage. Incorporating the system variability and improvement costs, the Generalized Process Of OnGoing Improvement (GPOOGI) is developed in this study. It shows that improving a frontend machine in a production line can be more effective than improving the throughput bottleneck. The findings indicate that we should consider the dependence among stations and the cost of improvement options during productivity improvement and should not simply improve the system bottleneck according to the conventional TOC. According to the GPOOGI, the managers of production systems would be able to make optimal decision during the continuous improvement process.