Modeling and analysis of output variability in discrete material flow production systems

Abstract

Developing analytical models for performance evaluation of production systems has been subject to numerous studies in the literature [4, 15, 24]. The main focus in most of these studies has been on utilizing Markovian models and deriving various first-order performance measures from the steady-state probabilities. The most commonly used performance measure in these studies is the throughput that is defined as the number of products produced per unit time in the long run. In addition average inventory levels, the average time spent in the system, probability of stock-out, probability of blocking and starvation are also used to design and control production systems by using these analytical models.