The 4PL Network Optimization Considering The Customer Behavior Under The Stochastic Demands

As the market competition intensifies,in order to improve their own competitiveness,enterprises begin to focus on their main business,and gradually begin to subcontract their logistics business to the professional logistics provider.As the supply chain integrator,the Fourth Party Logistics(4PL)has a high ability of supply chain integration,and could improve the operational efficiency of the logistics supply chain with the minimal logistics cost.So,researches related to the Fourth Party Logistics are of significance both in theory and in practice.The logistics network is bound up with the business of the Fourth Party Logistics.And logistics network optimization plays a crucial role in reducing the cost and improving the operational efficiency of the logistics.However,in different period of time,the demand of the enterprise is not fixed,which puts forward higher requirements for the Fourth Party Logistics network design.So,in this thesis,we study how to optimize the Fourth Party Logistics network with customer's behavior and stochastic demand in consideration and consequently improve the level of logistics operation.First of all,we portray the customer behavior with the value function of Prospect Theory,and build a mathematical model for the Fourth Party Logistics network optimization problem.We maximize customer's satisfaction under the constraint that the cost of the network satisfies the cost limit required by the customer.Secondly,according to the characteristics of the problem,we design an improved Particle Swarm Optimization algorithm,and propose adaptive inertia weight and adaptive learning factors.Then,we test the performance of the algorithm by three instances of different scales and acquire an algorithm parameter table of improved Particle Swarm Optimization algorithm.And through comparing the improved Particle Swarm Optimization algorithm with the standard particle swarm optimization algorithm,we verify the improved Particle Swarm Optimization algorithm is superior to the standard Particle Swarm Optimization algorithm.Finally,we analyze the 4PL Network Optimization considering the customer behavior under the stochastic demands.The results show that:1)with the increase of service level that the customer requires,the customer's satisfaction decreases;2)with the increase of the upper bound of cost that the customer requires,the customer's satisfaction increases.Meanwhile,we compare the situation which takes behavior into consideration to the situation which doesn't.The results show that under the situation considering behavior,the variation of customer's satisfaction is more sensitive.