The Design Of Distribution Plan Based On Stochastic Demand

Delivery process,including procurement,storage and distribution.In the last part of the delivery process,the random needs will occur from time to time,customer will generate new demand at any time,if you blindly follow the original delivery plan to continue,it is clear that there is no way to meet this random demand,resulting in reduced customer satisfaction.How to respond to customer’s needs quickly and accurately,providing delivery service to customers timely and satisfactory.It has become an urgent problem to be solved by distribution system.This paper studied how domestic and foreign scholars solved the relevant problems,concluded that there are two main ways to respond to random demand,one is focused on the study of customer’s demand forecasting,but did not optimize the route to achieve the goals of lowest cost;One is pay more attention on optimizing routes,but there is no reasonable prediction of the random demand,so there is no goods to meet the random demand immediately.In this paper,we establish a model of multi-objective programming according to the actual background of company.The model takes the total cost and the customer’s dissatisfaction as target value,Then three types of meta-heuristic algorithms such as ant colony algorithm,taboo search algorithm and genetic algorithm,whoes principle and adaptability are compared.Finally,a genetic algorithm,which is not easy to fall into the global optimum,parallelism and expansibility,is selected as the method to solve the distribution route.It is used to solve the initial distribution route of logistics distribution and adjust the planned route in real time after generating random demand.In the meantime,in order to deal with stochastic demand,we integrate the forecast of stochastic demand into the distribution plan,and compare the quantitative forecasting method,time series forecasting method and causal relationship forecasting method.Finally,gray model,which is more suitable for short-time forecasting and higher accuracy,is selected to calculate the proportion of goods,fully utilizing the redundant space for vehicles,and rationally carrying out more goods,so as to meet random demands efficiently and accurately.Finally,under the actual background of a distribution company in Dalian,the adjustment of delivery route at random demand is demonstrated,which proves that reasonable assembly and real-time route change can reduce costs and reduce customer dissatisfaction effect,and achieved good results.Also make greater practical value in the study of this article can be reflected.