New heuristic to generate an initial basic feasible solution for the balanced transportation problem

Abstract

This paper studied the balanced transportation problem. Many algorithms were developed to solve optimally this problem. The most commonly used method is the modified distribution method (MODI). The MODI algorithm starts by generating an initial basic feasible solution using the northwest corner method (NCM). In general, the cost of the initial basic feasible solution generated by NCM is too high and far away from the optimal cost. For this reason, we propose in this paper a new method to generate an initial basic feasible solution to the balanced transportation problem to solve this issue. This method, named Global Minimum Method (GMM), was tested and compared with the NCM, the minimum cost method (MCM), the Vogel's approximation method (VAM), and the optimal cost obtained by MODI. The computational results proved that the new heuristic is promising for small size instances and outperforms all other methods for big size problems.