A Genetic Algorithm with Priority-Based Encoding for the Multistage Supply Chain Distribution Problem

To support strategic decisions in logistics planning, we study a supply chain distribution problem close to real world situation. A basic assumption in any distribution problem is that the cost of transportation is directly proportional to the number of units transported. However, this assumption cannot be justified in real world situations. This paper described a distribution problem in which the transportation cost comprises a variable cost component and a fixed cost component. It is assumed that the variable component is continuous and linear increasing with volume of freight. The fixed cost incurs whenever there exists transportation between a source and a destination. The proposed problem can be formulated as an optimal mathematic model. A genetic algorithm with priority based encoding structure, is designed and applied, for solving the proposed model. At the end, a simple numeric example was introduced to apply the presented model and algorithm.