Performance analysis of a kitting process in stochastic assembly systems

Kitting (accumulating components required for an assembly) plays a crucial role in determining the performance of a small-lot, multi-product, multi-level manufacturing system. In this paper, we analyze the kitting process of a stochastic assembly by treating it as an assembly-like queue. Specifically, we investigate the dynamics involved in a simple kitting process where two independent input streams feed into an assembly process. Unlike previous studies in this domain, we relax the assumption of finite buffer capacity constraint on the input buffers, and still show that the system remains stable under fairly mild conditions. It is expected that the findings of this study will provide manufacturing system designers with wider variety of control parameters to choose from in evaluating the system performance under a much broader set of control policies, which would lead to minimizing the associated costs.