Towards performance analysis of manufacturing systems using nonlinear dynamic principles

Abstract

The objective of this work is to develop a modeling approach towards dynamic control and analysis of lean manufacturing environments. Traditional discrete-event system (DES) models tend to be time consuming and unwieldy for real-time applications. The new approach is based on nonlinear dynamic principles that can model complexity with real-time adaptability without the large computational time. The resulting model will help to track variations in the performance of a manufacturing system due to changing external and internal conditions. Based on the model, one could design/redesign appropriate inventory and supply management policies which form the essentials of a lean manufacturing system. Also, modeled characterizations are useful for analyzing alternative "what-if" scenarios and thus provide a comprehensive view of the ramifications of disturbances such as machine break downs or rush orders in the system. This paper explores the possibility of improving the performance and implications of modeling a pull-based manufacturing system using a new nonlinear dynamic systems approach. The details of the nonlinear modeling approach as well as its application to estimate the performance of a three-stage manufacturing system are presented. Advantages can be realized from savings in model execution time, faster rendering of performance estimates, and readily available analysis toolboxes that can become instrumental in implementing this new real-time nonlinear dynamic approach. Copyright © 2006 SAE International.