Optimal material release times in stochastic production environments

The problem of determining the 'optimal' material release times is an important and complex problem for most manufacturing firms. In this age of global competition, firms need to have low inventories, short cycle times and the ability to meet customer orders on time. In this paper, we consider a firm operating in a Make-To-Order environment and formulate this problem as an unconstrained cost minimization problem for a fixed sequence of work. We propose a solution methodology using stochastic approximation and infinitesimal perturbation analysis of a transient state simulation of the system. We discuss the implementation of this algorithm and present some preliminary results. We also present an algorithm to obtain a lower bound on the optimal objective function value under certain conditions.