An integrated control framework for flexible manufacturing systems

Abstract

This paper presents the design, development, and implementation of an integrated control framework that provides a real-time supervisory control model with limited look-ahead capability for flexible manufacturing systems. Control goals and policies are modeled and characterized by a fuzzy rule base, which is integrated with the control model. The framework consists of a finite state machine generator and a controller. The generator model is equipped with an output function and output sets. The controller model has a four-stage decision-making structure. The controller monitors performance measures of the manufacturing system and reacts according to the changes in the system states in order to keep the performance measures at desired levels. The integrated framework has been implemented on a software platform in order to validate its effectiveness. The performance of the framework has been tested on a hypothetical flexible manufacturing system using a simulation.

Application of the analytical hierarchy process for real-time scheduling and part routing in advanced manufacturing systems

“…These results are typical for not only manufacturing systems but for all other systems that are controlled by a control theory based system with a limited look-ahead horizon. The results are also consistent with Buyurgan and Mendoza [6] and Buyurgan and Saygin [7]. For N = 1 and N = 2, the scheduler model makes myopic estimations and shallow decisions for such a short horizon due to its very limited visibility of future states.…”

“…This paper extends the research conducted in Buyurgan and Mendoza [6] and Buyurgan and Saygin [7]. In those studies, dynamic control of manufacturing systems is investigated using extended finite state machines.…”

“…The presented model in this study is the extended finite state machine model, proposed in Buyurgan and Mendoza [6] and Buyurgan and Saygin [7]. The first study uses the presented finite state machine model and utilizes the preemptive goal programming method where performance measures are prioritized and future states are ranked accordingly.…”

“…The dynamic manufacturing system scheduling model presented in this paper is extended from the work of Buyurgan and Saygin [7] and Buyurgan and Mendoza [6]. The model is based on optimization of desired performance measures.…”

“…These states are ranked based on the predetermined priority of performance measures, and the state with the highest ranking is selected as the preferred future behavior. In Buyurgan and Saygin [7], uncertainty and vagueness are introduced to the future-state assessment process. Performance measures are represented in fuzzy sets to incorporate the uncertain and complex correlation among different criteria into the decision-making process.…”