Expected value model of bus gas station site layout problem with fuzzy demand in supplementary fuel using genetic algorithm

In view of the imbalance of vehicle refueling needs belonging to different routes in time and space, this paper presents a multi-objective optimization model to design the site layout of bus gas stations, in which a fuzzy vehicle fueling demand for each line is considered. The proposed model features the process of vehicle refueling in bus gas stations abstracted as multi-server queuing systems to analyze the influence of the location of fueling stations on bus scheduling. Based on credibility theory, the expected value model of the site selection for bus gas stations is established, where some practical factors such as vehicle travel, queuing and refueling are comprehensively considered. The primary objective is to minimize the construction cost of filling stations, while the secondary objective is to minimize the cost of refueling all buses. According to the characteristics of the problem, it is presented as a deterministic, single objective, linear programming model. Genetic algorithm is designed to solve this problem by defining the coding scheme of solutions, fitness function, and the heuristic algorithm of generating the initial population. Finally, the optimal site selection scheme for bus gas stations is calculated and validated through a numerical example. The influence of fueling station’s ability on its layout is analyzed, thereby the validity of the model and algorithm is verified.