Research On Fourth Party Logistics Routing Problem Based On Stochastic Programming

In the current research on the fourth party logistics routing problem,most of them stay in the research stage of deterministic problems,and do not consider the impact of uncertain factors such as seasons and human operations on actual transportation,but these uncertain factors will lead to deviations in logistics transportation to some extent..Based on the fourth party logistics path optimization,stochastic programming,combined transportation and intelligent optimization algorithms,this paper focuses on the randomness of transportation time,transportation cost and loading volume caused by various uncertainties in logistics transportation,and The fourth-party logistics path optimization problem was studied considering the characteristics of different transportation modes with corresponding technical and economic characteristics.The specific research contents are as follows:This thesis researchs on the fourth party logistics routing problem considering stochastic demand and multiple transportation modes.Due to the changes in the habits of the current consumer population,the customer’s demand for goods presents a certain regular shape and pattern.In order to meet the customer’s demand and ensure high quality service,the fourth party logistics integrator integrates the distribution service information of the third party logistics supplier and the customer’s demand,and makes overall planning for the distribution task,and develops a reasonable logistics distribution plan.Thence,considering that in the fourth party logistics,customers demand is characterized by timeliness and regularity,and multiple transportation modes provided by the third party logistics providers in reality,the problem of fourth party logistics routing problem with stochastic demand and multiple transportation modes are studied.Given that customer demand is a random variable,the chance constrained program model is designed with the goal of minimizing transportation costs,which determines the transportation modes while selecting transportation path and logistics 3PL providers.To solve the model,an improved genetic algorithm with embedded migration operator and elite strategy is designed.The algorithm parameters optimized by Taguchi test effectively improve the accuracy of solutions.The experimental results show that with stochastic demand of customers,the transportation cost increases with the different confidence level.And joint transportation through multiple transportation modes can overcome the defects of single transportation mode and effectively reduce the transportation cost.The proposed algorithm has better global search capability and computational accuracy,and the IGA can solve the problem effectively.Based on the first study,this paper also studies the fourth-party logistics path considering random transportation time and transportation costs.Considering the uncertainty of transportation time and transportation cost during transportation due to seasonality and human factors,a multi-objective opportunity constraint planning model with minimum transportation time and transportation cost was established.According to the customer’s emphasis on transportation time and transportation cost,the objective function of multi-objective opportunity constraint planning is normalized.According to the characteristics of logistics transportation network path optimization problem,the hybrid beetle swarm optimization algorithm embedded in Dijkstra algorithm is designed to solve the problem,and case analysis and algorithm analysis are carried out for three different scale cases.By adjusting the model parameters,the minimum objective function value under the combination of parameters is obtained,and the influence of the customer on the transportation time and transportation costs is different and the impact on the whole logistics and transportation scheme is obtained.The Taguchi test was used to determine the optimal parameter combination of the mixed Astronomical herd algorithm under different node scales.The beetle antennae search algorithm,the genetic algorithm and the particle swarm optimization algorithm embedded in Dijkstra algorithm were compared to verify the optimization of the hybrid beetle swarm optimization algorithm in the fourth party logistics path validity and feasibility.