Traveling salesman problem approach for solving drug distribution using simulated annealing and Tabu search

Abstract

In combinatorial mathematics and operational research, the traveling salesman problem is among the most studied problem which aims to find the least possible cost or distance when visiting all the cities exactly once. This research analyzes a drug distribution problem, which is regarded as an application of the traveling salesman problem. Two meta-heuristic methods were utilized to generate the best solution, which are the simulated annealing and tabu search. There are essential parameters for both of the methods. In the simulated annealing, the right setting for the cooling schedule is very important so that the algorithm will converge to a near-optimal solution. Moreover, the size of the tabu list is an essential value for the tabu search method since it will determine how the algorithm will search for better solutions in the search space. The results of this experiment were analyzed based on the output generated by the developed C++ program, and it revealed that the simulated annealing and tabu search methods were capable of generating a best solution within a short computational time.