قاعده نسل فازی برای برنامه ریزی تطبیقی ​​در محیط تولیدی پویا

This paper proposes a fuzzy rule-based system for an adaptive scheduling, which dynamically selects and applies the most suitable strategy according to the current state of the scheduling environment. The adaptive scheduling problem is generally considered as a classification task since the performance of the adaptive scheduling system depends on the effectiveness of the mapping knowledge between system states and the best rules for the states. A rule base for this mapping is built and evolved by the proposed fuzzy dynamic learning classifier based on the training data cumulated by a simulation method. Distributed fuzzy sets approach, which uses multiple fuzzy numbers simultaneously, is adopted to recognize the system states. The developed fuzzy rules may readily be interpreted, adopted and, when necessary, modified by human experts. An application of the proposed method to a job-dispatching problem in a hypothetical flexible manufacturing system (FMS) shows that the method can develop more effective and robust rules than the traditional job-dispatching rules and a neural network approach.

Optimizing production scheduling is one of the most important ingredients for high productivity in modern manufacturing industries. Traditional approaches to solve the scheduling problems can be classified into three categories: analytical, heuristic, and simulation approaches [1]. The analytical approach uses mathematical programming models, stochastic models, and control theory. However, it is applicable to only small-sized problems because of NP-completeness inherent to the scheduling problems [18] and [23]. To overcome the mathematical difficulties, heuristic approaches have often been adopted for efficiency at the cost of optimal decision. Most heuristics have focused on dispatching rules that determine the dispatching priorities of machines, automatically guided vehicles (AGVs), or jobs. While many dispatching rules have been proposed and evaluated, it remains hard to prove the general usefulness of a rule in spite of various system characteristics.

This study proposed a new method of adaptive dispatching in a typical dynamic manufacturing environment. The method includes an automated knowledge acquisition scheme to build the fuzzy rule base for the adaptive scheduling. A dispatching decision is to pick up a job among competing candidates that are waiting to be processed. The decision is made on the basis of selected criteria with which the urgency of each job is represented and the weights for the criteria. The weight vector, which represents the decision-making strategy, is varied for optimum performance according to the current system state. The knowledge acquisition of the optimal mapping between the system states and the appropriate weight vectors is automated with a computer simulation of the system. In this study, the decision-making is made by combination of criteria with optimized weight sets in different system states. Simulation results showed that the proposed method produced highly productive and robust schedule than conventional heuristics in the hypothetical manufacturing system.