Multi-criteria optimization of multi product assembly line using hybrid Tabu-SA algorithm

Abstract

In this paper, a multi-criteria optimization problem has been addressed including minimizing three important objectives: (a) total amount of fines for untimely production of orders; (b) total carrier rails' time including the preparation times of the stations; and (c) total cost of feeding outside the work station plan. In order to consider the relative importance of the stated goals, all costs/fines have been weighed. The proposed problem is one the nondeterministic polynomial complex problems, for which no comprehensive solution has been presented. Hence, to find the optimal answer in a reasonable time, Tabu search (TS) and simulated annealing (SA) algorithms have been used. Furthermore, to obtain the merits of the stated algorithms, a combination of both has been employed (Tabu-SA algorithm). To illustrate the performance of the proposed method, two numerical examples have been solved. The computational results indicate the high speed of integrated search algorithm and the superiority of the quality of its solutions. Based on the results, the proposed hybrid Tabu-SA algorithm has a remarkable advantage in comparison with TS and SA algorithms in terms of convergence rate. The total cost has been reduced at 24% using the proposed hybrid Tabu-SA algorithm. Besides, the computing time was improved at 29% and 46% using the proposed procedure respectively.Keywords Multi-criteria optimization · Multi modal assembly lines (MMAL) · Tabu search (TS) algorithm · Simulated annealing (SA) algorithm * Ahmad Tavakoli, tavakoli-a@um.ac.ir |