An extended location model, a single-source continuous capacitated facility location-allocation problem (SSCCFLAP), has been developed. Since this problem has conflicting objectives in nature, it is impossible to find the solution directly, thus heuristic methods were introduced to approach the problem. Heuristic approaches are very efficient, however, they cannot guarantee global or near global optimum solutions. In order to find global or near global optimum solutions, global optimum search methods, such as evolutionary algorithms, were applied. Because of the time consuming process of evolutionary algorithms, hybrid computational methods were applied. The studied methodology, which is based on evolutionary algorithms and the integration of a modified non-identical bin-packing heuristic method, can approach global, or near global optimum solutions, and still maintain its efficiency. The studied method has been implemented in Visual Basic and the system has been verified and validated. In conclusion, the results of this research present the beginning of an extended location model that is closer to the realistic situation, and also provides a hybrid evolutionary algorithm approach.
展开▼
