Track: Supply Chain Management

Abstract

The research study presents a developed calculation algorithm for delivery sequencing for real-time purchasing orders in LPG (Liquefied Petroleum Gas) delivery business in Thailand. In practice, customers call the LPG call center requesting services and each customer’s order mostly requires an immediate service. Then routing and scheduling is done in real time. That the delivery provider needs to make a decision all the time due to the incoming orders are increasing making things more complicated in routes and schedules. As a result, the delivery cannot determined well enough in an appropriate time period. Hence, the objective of this study is to develop a sequential delivery calculation algorithm for LPG Delivery in real-time continuous incoming purchasing orders for calculating an appropriate total distance and reducing lead time before dispatching. This algorithm has been developed for sequencing and routing a fleet of vehicles from a distribution center (DC) to multiple delivery points. The vehicle routing method applies heuristics techniques on vehicle capacities and distances to minimize the travel distance and do the sequencing by on a real time basis. The system is divided into 3 phases. The first phase is to collect the gradually incoming data for calculation. As soon as a customer orders LPG, the system uses the Google MAP API for generating a distance data matrix to calculate the distance between customer’s location and the nearest depot. The second phase is order assignment under the restriction; vehicle capacity and distances between routes and mapping out an appropriate route by using Nearest Neighbor Heuristics and 2-opt improvement. Finally, the third phase is to schedule the dispatches. The significance of the sequencing system is that it is all executed in real time. The findings in the research can be used by integrating the system to help the involved users in reducing the decision-making time for efficient routing, leading to less lead time before dispatching goods, having an efficient fleet control, and a right in time delivery with an appropriate total distance. To summarize, the system provides significantly better routes than those determined manually.