Font Size: a A A

Service Optimization Methods For Distribution Network In Supply Chain Management

Posted on:2013-06-06Degree:DoctorType:Dissertation
Country:ChinaCandidate:L N ZhouFull Text:PDF
GTID:1228330392467539Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Distribution is an important part of supply chain management. Distributiondecision and optimization have attracted considerable attention from bothpractitioners and academics. Most of the previous researches focus on costreduction to improve the enterprise’s competitiveness. Recently, with thedevelopment of productivity and information technology, the degree of producthomogeneity is high and the market is currently a buyer’s market. Customers intoday’s marketplace are more demanding, not just of product quality, but also ofservice. Cost reduction is no longer the magic weapon to compete successfully, andthe enterprises come to realize the importance of customer service. In this regard, awidely used manufacturing distribution system is investigated in this thesis, sometypical distribution decision and the service optimization methods are proposedbased on operational research and artificial intelligence theory. The maincontribution of this thesis can be summarized as follows:Firstly, the connotation of distribution service is expounded, the distributionservice level evaluation system is set up, and the main Influence factors areanalyzed. Then a widely used manufacturing distribution system is investigated.Based on its service optimization requirements, the research problems are presentedand the optimization rules are analyzed. Concerning the common characters of theseproblems, the differentiation based optimization strategy and the multi-objectiveoptimization methods are presented. Finally, a service-oriented optimizationsolution for the distribution system is proposed as the research foundation.The service-oriented optimization method for redesigning distribution networkwith increments is studied. Weighting factors are employed to strengthen theinfluence of the important customer regions and products, and a new method isproposed to calculate the unitary service level of the distribution system. Based onthe fuzzy analytic network process, the weighting factors are calculated and then amulti-objective optimization model for distribution network reconfiguration isproposed. Concerning the character of the model and the complex nature of thesolution structure, a determinant code-based encoding structure is designed torepresent the problem, and a genetic algorithm-based solution procedure is developed. Simulation experiments and the comparisons are conducted to show theeffectiveness of the proposed method in improving unitary service level, keepingimportant customer’s service level, and reducing distribution cost.Considering the optimization requirements of reducing the retailer’s storagecost and improving customer service level, the single-product inventory systemwith stochastic demands and random replenishment intervals is studied. Based onthe service differentiation principle, inventory control point is introduced to ensurethe important customers’ priority, and a new inventory control strategy based oncustomer segmentation is proposed. Then M/M/1queue system is established todescript the random inventory process. The inventory indices are calculated by thesteady state analysis, and the integration inventory optimization model based oncustomer segmentation is proposed. To deal with multi-objective and incorporatethe decision maker’s preferences in the decision process, goal set and relaxationitems are employed and the goal attainment method is used to solve the problem.Finally, simulation experiments and the comparisons are conducted to show theeffectiveness of the proposed method in reducing inventory cost and improving theon-time service rate.To cope with the shortage problem of the sub distribution system, operationoptimization method is studied. Retailer differentiation is distinguished byintroducing service level threshold, an integrated lateral transshipment policy basedon service level adjustment is proposed, and a multi-objective mathematical modelis formulated to optimize the distribution operation process. To calculate thetransshipment policy variables, and to ensure the stock-out retailers’ priority, themulti-objective memetic algorithm with priority-based encoding is developed.Finally, simulation experiments and the comparisons are conducted to show theeffectiveness of the proposed method in controlling transshipment cost, reducingservice delay rate, and shortening customer waiting time.A service-oriented supply chain distribution management system is designedand implemented based on the strategies, models and algorithms proposed in thisthesis. Combining with the real problems in distribution management of HarbinHafei Automobile Industry Group Company LTD, the simulative application andresults are given to validate the theory and methods proposed in this thesis.
Keywords/Search Tags:supply chain distribution management, service optimization, differentiation, distribution network optimization, inventory decision, lateral transshipment
PDF Full Text Request
Related items