Variable neighborhood search heuristic for the full truckload problem in liquefied petroleum gas supply

In this paper, we propose a generalized version of the full truckload vehicle routing problem FTVRP in the liquefied petroleum gas (LPG) supply chain. We focus on the supply component that is considered to be the main element of LPG supply chain. The objective consists of determining a schedule of weekly supply which minimizes as well the gas acquisition cost as that of its transport. In order to effectively solve this problem, the mathematical model is established and a neighboring search solution heuristics (VNS) is proposed. The initial solution is generated using a Greedy Algorithm by selecting the most beneficial nodes in terms of transportation cost and gas acquisition cost. In the experiment, different sizes' problems are used and the numerical results getting with big instance are explored.