An Investigation of Stochastic Analysis of Flexible Manufacturing Systems Simulation

This paper describes the use of an ARIMA (autoregressive-integrated- moving-average) model and its equivalent state space models to produce rule-based knowledge for flexible manufacturing systems (FMS) that can be used to investigate a wide variety of problems including machine breakdown, material shortage, and changes of scheduling rules. One great advantage of using the proposed models is the ease with which the simulation results can be summarised, analysed and captured, as well as the availability of the mathematical representation of the knowledge that can be kept in a knowledge database for evaluation and selection of alternative FMS strategies in a real-time environment. Various case studies are used to illustrate the methodology and the development of ARIMA and state space models, the analysis includes the system cost and stability of changes or interventions, the relationships among the simulation inputs and outputs, and the formulation of the production rule-base for the FMS scheduler. Management can use this integrated approach to describe and predict the dynamic behaviour of a complex FMS.