Short-term Coastal Inventory Routing Problem Based On Coal Transportation

Coal is an important strategic energy source in China with huge demand and ever-increasing demand.Over 50%of the coal is used for thermal power generation.Fuel coal is the type of coal used to generate electricity for power plants.However,coal production in China is concentrated in the northwest,but consumption concentrated in the central and southern regions.The pattern of energy distribution leads to the important role of maritime transportation in the distribution of coal resources.With the change of national policies,market demands and transformation of industrial structure,a production and marketing integration pattern has been formed,the pattern makes the research on the transportation and inventory of coal possible to draw lessons from the research method of marine inventory-routing problem.As a result,the study on fuel coal inventory-routing problem is of great significance.I hope this article can provide reference for future research.Based on the analysis of previous literature on maritime inventory-routing problem and the full consideration of the characteristics of fuel coal transportation and inventory,this paper studies a limited planning cycle,many-to-many topology,heterogeneous fleet and coastal short-term fuel coal Inventory-Routing Problem.The purpose of this paper is to determine the inventory policy of fuel coal in power plants and the transportation plan,under the constraints of meeting the power generation demand of the power plant and the inventory restriction.First of all,a path-flow model is established with the lowest total operating cost of coal transportation and storage.Then,according to the characteristics of the model,an approximate Lagrangian relaxation algorithm is proposed-After that,the original problem model is transformed and decomposed and a sub-gradient method is used to iteratively approximate the optimal solution.Finally,taking a power generation group as an example,the model verification and solution analysis are carried out to further verify the model and the feasibility of the algorithm.