The Minimum-cost Flow Algorithm With Multi-warehouse And Multi-warehouse Constained Route For Uncertain Network

The Festival of Double Eleven is an artificial festival at 11 th Nov,when a great deal of companies conduct big promotions,and thus a huge amount of orders are placed by the customer that day.Although it’s a great pleasure for the seller to receive a lot of orders,it’s not that easy to deliver the goods,due to the complexity of the scheduling of the transportation,which comes from the uncertainty of the availability of goods,the transportation environment,etc.Here,the uncertainty theory may be leveraged to help us to find a schedule that both deliver the goods to the customer as soon as possible and have a minimal logistics cost.Cargo transportation scheduling problem(Referred to as transportation scheduling)is a decision process of allocating the limited resources to a variety of different needs,aiming at optimizing one or more goals.Since the transportation scheduling problem is general among today’s manufacture and production system,a more optimized transportation scheduling improves the delivery efficiency of goods,reduces the cost of logistics,and thus,leads to better economic benefits and market competitiveness.Therefore,it is of great importance to carry out an effective and reasonable scheduling.Based on the analysis of the existing researches,this paper studies the minimum cost transportation scheduling problem of multi-warehouse-and-multi-variety transportation path on an uncertain graph.Although it is possible to conduct a complete search to find the planning solution,its huge search space(and thus high time complexity)is simply unacceptable(not to mention the unnecessary waste of computing power).In this paper,we try to design a two-step algorithm,which should have a far smaller search space,based on the uncertain theory and uncertain planning theory.Firstly,we try to design an algorithm that work out the reachability relationship between the states in the search space by leveraging the properties of directional loops,taking into account uncertainties such as the transportation environment.To be clearer,we use a directed graph to represent the state space,and try to find cycles inside it.Since cycles in graphs signify mutual reachabilities of the nodes on the cycle,we may treat the entire cycle as a single node in the search space,and thus reduce the search space.Secondly,according to analysis of the real situation of multi-storey multi-warehouse with the restriction of transportation path,we designed the minimum cost transportation algorithm with multi-warehouse and multi-warehouse transportation route,which gives the minimum cost schedule.This algorithm runs on the graph produced by the previous algorithm,and tries to make plans for different categories of goods one by one.Each select a delivery path,the increase of the uncertain environmental factors of P,and each path have a critical value of plimit limit to ensure the safe delivery of goods.To minimize the total shipping cost,the algorithm greedily chooses the least cost edge,with SLF/LLL strategy as an optimization.Once a category of good is out of stock in some warehouse,the algorithm checks if a more optimized solution is found,and adjust the final result accordingly.