联合采购策略下的选址—库存—配送协同优化模型与智能算法研究

发布时间：2018-03-15 05:15

本文选题：果蝇优化算法　切入点：联合采购　出处：《华中科技大学》2016年硕士论文　论文类型：学位论文

【摘要】：为有效降低成本,跨境电商企业需要尽可能采用大批量、少规格的方式从国外进口商品,这与消费者小批量、个性化、实时化需求之间的矛盾非常突出。为缓解这一矛盾,有效降低成本的同时还能快速满足消费者的需求,建立配送中心并保持合理的库存就显得尤为重要。联合采购(Joint Replenishment,JR)策略是一种有效的成本控制手段,而在配送中心建设运作中,选址、库存和配送(location,inventory,and delivery,LID)都是非常关键的决策要素。基于JR策略考虑配送中心选址、库存、配送(JR-LID)协同优化研究应用价值更高,但是设计高效、稳定的高精度算法非常有挑战。本文设计了一种改进的果蝇优化(Improved Fruit Fly Optimization,IFOA)算法,并用于求解配送中心JR-LID协同优化难题。首先,对基本的果蝇优化算法进行两点改进,在算法视觉搜索阶段,利用果蝇的群体协作,选择具有更好适应度值的果蝇个体朝向新位置飞行,而适应度值差的果蝇个体在搜索空间中随机飞行;此外,加入一个随机扰动因子,如果本次迭代的最优解比迄今为止已发现的最优解差,则采用扰动的方式改变对应果蝇的位置,以跳出局部最优解。通过18个标准测试函数验证了IFOA的性能。其次,将IFOA用于求解典型的联合采购问题(JR problem,JRP),该问题为NP-hard问题。对比算例结果证实明IFOA获得的结果优于目前最好的智能算法。最后,讨论了在三阶段供应链网络下考虑JR策略的配送中心“选址-库存-配送”协同优化模型求解问题。设计基于IFOA的求解算法,从而确定优化的配送中心数量、建设位置、联合订货策略和配送作业方案。通过算例和敏感性分析可为企业配送中心运营管理提供决策参考,同时通过扩展算例的求解进一步验证了IFOA算法的有效性和稳定性。
[Abstract]:In order to effectively reduce costs, cross-border e-commerce enterprises need to import goods from abroad in as large a quantity and less specifications as possible. This contradiction with consumers' small volume, personalized and real-time needs is very prominent. It is very important to set up distribution center and maintain reasonable inventory while effectively reducing cost and meeting the needs of consumers. Joint Replenishment JRR is an effective means of cost control, but in the construction and operation of distribution center, JRR is an effective means of cost control. Location, inventory and distribution inventory and delivery LIDs are all very critical elements of decision making. Considering distribution center location, inventory, and distribution JR-LID-based collaborative optimization research is more valuable, but the design is more efficient. The stable high precision algorithm is very challenging. In this paper, an improved Fruit Fly optimization algorithm is designed, which is used to solve the problem of JR-LID cooperative optimization in distribution center. Firstly, two improvements are made to the basic Drosophila optimization algorithm. In the visual search phase of the algorithm, the individual flies with a better fitness value are selected to fly towards a new position, while the individuals with poor fitness value fly randomly in the search space, using the collaboration of the drosophila population. If a random perturbation factor is added, if the optimal solution of this iteration is different from that found so far, the position of the corresponding fruit fly will be changed by perturbation. In order to jump out of the local optimal solution, the performance of IFOA is verified by 18 standard test functions. Secondly, IFOA is used to solve a typical joint procurement problem, which is a NP-hard problem. The comparison of the results of an example shows that the results obtained by IFOA are superior to those obtained by the best intelligent algorithm. This paper discusses the solution of the "location-stock-distribution" collaborative optimization model of distribution center considering Jr strategy in three-stage supply chain network. A solution algorithm based on IFOA is designed to determine the number of optimized distribution centers and the construction location. Through the example and sensitivity analysis, it can provide the decision reference for the enterprise distribution center operation management. At the same time, the validity and stability of the IFOA algorithm are further verified by solving the extended example.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：F274;F224

【参考文献】