سال انتشار: ۱۳۹۰
محل انتشار: همایش منطقه ای پژوهشهای نوین در ریاضی
تعداد صفحات: ۶
Rassoul Noorossana – Ph.D., Professor Department of Industrial Engineering, Iran University of Science and Technology, Tehran, Iran
Moslem Toosheghanian, – M.Sc. Student Department of Industrial Engineering, Iran University of Science and Technology, Tehran, Iran
In recent years, it has been revealed that the control charts operating with VSI (variable sampling interval) schemes give better performance than with FSI (fixed sampling interval) schemes in the sense of quick response to process shifts. The economic design of a VSI control chart involves minimization of a complex nonlinear cost function that formulates the cost of implementing the VSI chart. In this paper, four metaheuristic algorithms are employed to find optimal values for the design parameters. These algorithms are as genetic algorithm (GA), particle swarm optimization (PSO), simulated annealing (SA), and ant colony optimization (ACO).To evaluate the algorithms, for each algorithm four criteria corresponding to obtained solutions are derived. The concerned attributes are (a) the expected loss-cost generated in the production cycle, (b) type I error or false alarm rate of the control chart, (c) test power or failure-detection power of the control chart, and (d) average time to signal (ATS) when assignable cause occurs. Rankings of the algorithms are determined by TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method. Obtained results show that in all cases the VSI control chart has a lower expected loss cost than corresponding FSIcontrol chart. Also, PSO and GA have presented better values for attributes than SA or ACO in solving the model.