A fleet scheduling optimization model: A case of natural gas distribution system in Thailand

This research presents a development and implementation of a web-based fleet scheduling system to support logistics process of the natural gas commerce in Thailand. The logistics network for vehicle-use natural gas in Thailand consists of two types of station: main stations and substations. Natural gas tanks, which can be refueled only at the main station, are delivered by trucks from the main station to its substations. This logistics problem can be classified as the hub and spoke distribution system with the main station as hub and the substations as spokes. The distribution process of natural gas has to be managed precisely and efficiently. A web-based system was developed to monitor logistics process and help planning and scheduling the fulfillment process. A new mathematical model is developed to create delivery schedules. The model is a mixed integer nonlinear programming model with the objective to minimize the total transportation costs. From numerical results, the new model shows more accurate predictions of gas shortages and can improve the performance of natural gas distribution system. The shortage time at the substations decreased significantly. Currently, the company is implementing this web-based system to manage the replenishment process of all of its 484 substations.