Research On Modern Logistics Some Problem

The most important aspect of modern logistics is the efficient organization, which is reflected in each of the management stages. The logistics management system including intelligent optimization algorithms has the advantages of high efficiency, low cost, reasonable consumption, and strong competition and business development capacity. Vehicle routing problem (VRP), logistics station selection problem and intelligent algorithms of this industry have becoming the hot spot of modern logistics research.This paper in-depth researches and implements the VRP, logistics station selection and some optimization algorithms based on the computational intelligence theory combined with classic logistics system instances, and presents the novel optimization algorithms to solve the practice problems. The works of this paper are as follows:(1),This paper presents the model of VRP first, and then constructs the improved algorithm based on computational intelligence method. Mainly, the model employs genetic algorithm to create heuristic algorithm, which is used in VRP with time window. The algorithm constructs the mathematic model, and then presents the novel chromosome selection policy and crossover operator to create solution. This algorithm solves 100 clients Solomon problem using the strategy of classification after the first sort. The validity and science are validated by the experimental results and practice application of decision-making system. The run time of algorithm is within one second applied to small-scale problem, ten seconds to medium-scale and ten minutes to large-scale problem. The percent of optimal result achieves 95%. The heuristic function of this genetic algorithm calculates the distance and time window factors; the deviation is very small between the result and known best solution and the solution path is built faster. This theory and algorithm has been applied into the application of Modern scientific Logistics Decision Support System.(2),This paper presents the improved genetic algorithm based on the Rank method to solve TSP problem with time window. The validity is proved by experimental result, this algorithm can construct the Pareto solution border too. The selection method based Rank is a solution independent of fitness value. It sets the different Rank value to the active and passive individuals. The algorithm does not set the fitness to two irrelevant individuals, which can give them more competition in the context of multi-object. The solution of multi-object constructed by this method is better, compared with others, this method provides the better service for the logistics service providers and provide guidance to the logistics needs of users.(3),This paper presents the mathematic model of logistics station selection, gives the station number and locations by quantitative analysis method. A genetic algorithm is presented to this decision-making process. This algorithm focus on the effective logistics path, namely logistics area, station location rationalization and optimization problem. It will confirm the nodes of logistics distribution network, goods and related resources preparation, cover area of each node, deliver class and method. The experiment runs with the different objective function according to the different problems. Compared with the actual condition, the results are the feasible solution, which prove the validity and correctness. This work is integrated into Modern scientific Logistics Decision Support System, the practicability is confirmed.(4),This paper first proposals and studies in depth the intelligent methods of china modern logistics development. We use some intelligent methods (multi-agent model based on belief) to solve problem model in the development of transportation system. A multi-agent system model is improved and then an approach described by belief is presented. The object of multi-objective optimization is to find a pareto optimal solution, which can be used in decision-making. Agent individual has powerful autonomy, it can provide the complicated adjustment policy. The model extends the function of belief and adds special rules, innovations and ancestor knowledge of different kind problem, which enhances the science and practicability of system. This paper summarizes the characteristics of core intelligent algorithms, researches and expatiates the classic application and those influences. Combining ITS system and logistics actual application, we summarize the influence of intelligent methods to the development of modern logistics system. The work makes the intelligent methods as the data and technology core of intelligent transportation system.(5),The actual application conditions of the works in this paper are reflected by Modern scientific Logistics Decision Support System. Moreover, the other three systems in different logistics fields are introduced, which presents the applications in other fields.