Research On The Green Logistics Capacitated Vehicle Routing Problem Based On Parallel Ant Colony Algorithm

For the logistics enterprises, the increasingly rising of fuel’s prices hasgreatly hurt its profitability and the healthy development of the industry.For the environment, the greenhouse gases and serious pollution generatedby the combustion of fossil fuels are getting worse and worse. For China,the energy supply is heavily dependent on imports and has no voice in fuelprices settings of the international market. From the perspective ofenvironmental protection, national energy security strategy, businessoperations, it is very important for us to reduce the fuel consumption of thevehicles in the logistics and transport and achieve sustainable economicdevelopment.This paper cut in the current issues from VRP perspective, hoping toreduce the fuel consumption in the logistics and distribution processthrough better path planning, because the most fuel-efficient pathes moreoften than not the shortest path. The innovation of this paper is toestablish a mathematical model of the actual situation, so as to reducefuel consumption under CVRP and VRPSDP, and find the connectionbetween fuel consumption factors of internal combustion engine andmathematical model of CVRP (VRPSDP), enabling theC VRP andVRPSDP model can represent engine’s fuel consumption quantitatively.Then we work out the solution with ant colony algorithm which has beenimproved by modern heuristic method. After the research of roadspecifications and construction standards in China, the internalcombustion engine fuel technology, we take running distance, cargoweight, road pavement， road grade into specially designed heuristicfactors. We give different priorities to different according to the stages,and make reasonable parameters setting according to the Chinese roadspecification as well as the vehicle’s actual fuel consumption situation. We differ from before in the pheromone update by learning from parallelant colony algorithm to better mimic the parallel strategy of ants in a realnatural environment, with reference to other the pheromone update rules.Finally, we further extend the model and algorithm to a simultaneouslypick-up and delivery conditions VRPSDP.As the ACS’s performances are sensitive to the parameter settings, adetailed parameter settings discussion were conducted afterwards. Finally,the experimental results obtained by the Matlab simulation show thateach heuristic factor makes contribution to save fuel around9%under theminimum fuel consumption targets. Thanks to VRPSDP return constraintsettings, it achieves lower fuel consumption. Although the length of themost fuel efficient path is longer than the minimum path by10%-20%,but around30%fuel can be saved. The experimental results verify thereasonableness of the assumptions of this thesis and the effectiveness ofthe algorithm, provides a new perspective of the VRP, and a direction ofhow to make the logistics become more green.