Due-Time Performance in Production Systems with Finished Good Buffers: A Systems Approach

点击翻译按钮获取中文摘要

In production systems with unreliable machines, the number of parts produced during a fixed interval of time is a random variable. Its distribution characterizes both the production volume and the productionvariability. The production volume has been analyzed in many publications. In contrast, the production variability has been studied in just a few recent articles where mostly production variance has been addressed. Although a useful measure of performance, it offers only a general, theoretical characterization of the production variability. Indeed, knowledge of the variance does not yet specify the probability with whicha fixed shipping schedule is satisfied. Recently, a new variability measure, referred to as the Due-Time Performance (DTP) and defined as the probability to satisfy a shipping schedule, has been introduced. The goal of this project is to develop methods for analysis and design of production systems from the point of DTP and to apply these methods in the automotive industry. Specifically, the following analysis problem will beaddressed: Given a production system, develop a constructive, analytical technique for DTP evaluation. Two design problems will be investigated. Direct problem: Given the reliability characteristics of the machines and the customer requirements on the shipping interval and shipment size, design a product delivery system so that DTP requirements are satisfied. Inverse problem: Given the customer requirements and the product delivery system, determine reliability characteristics of the machines required to satisfy a given DTP. If successful, this project will provide rigorous engineering tools for analysis and design of production systems in lean manufacturing environment operating in the pull regime. Their implementation will leadto a possibility of "calculating" a product delivery system rather than selecting it on the basis of ad hoc techniques in use today.