基于改进蚁群算法的应急物流车辆路径问题研究

发布时间：2018-04-17 06:43

本文选题：应急物流 + 中心选址　；参考：《济南大学》2015年硕士论文

【摘要】：车辆路径问题是当前应急物流领域研究中的热点问题,对突发事件的应急救援起着重要作用。国内外专家学者的研究主要集中在应急物资中转站选址和配送路径优化两个方面。本文结合实际情况探讨了灾区内应急物资临时中转站选址与路径优化集成解决的问题,分别构建了在静态与动态需求条件下的中转站选址与路径优化模型,并根据构建的模型设计了改进的蚁群算法进行求解,算例分析和实证研究验证了模型的有效性与算法的优越性。研究结果对于突发事件发生后应急管理部门的及时应对和有效决策具有重要的理论价值和现实意义。本文的研究主要分为以下几个部分:第一部分,绪论。主要阐明了应急物流车辆路径问题研究的背景和意义,并在对国内外研究现状进行综述分析的基础上,总结了当前该领域的研究成果和进一步研究空间,提出了本文的研究方法、研究路线及主要的创新之处。第二部分,应急物流车辆路径问题相关理论概述。结合我国在应急物流领域的相关研究现状,对研究所涉及的应急物流、车辆路径问题、蚁群算法、动态规划等理论进行了概括分析,进一步完善和充实了理论基础。第三部分,静态需求条件下应急物流中心选址及路径优化。主要研究了带软时间窗的多供应点多需求点多物资需求的应急物流车辆路径问题。首先,引入应急需求最大化覆盖思想和时间、成本最小化思想,集成构建了应急物资中转站选址与路径优化的多目标规划模型;其次,针对构建的多目标规划模型,结合多种智能算法思想,在信息素更新与限制、禁忌表共享、状态转移概率等方面对蚁群算法进行了改进,提高了算法的收敛速度和求解质量,并对适用于相应模型的改进蚁群算法进行了设计;最后,通过算例分析验证了模型的有效性,并通过对比分析显示了改进算法的优越性。第四部分,动态需求条件下应急物流路径优化。在静态路径优化模型的基础上引入动态规划思想,对比分析静态路径优化问题与动态路径优化问题的特性,提出了本文要解决的动态路径优化问题;结合实际情况,重点考虑了出现新的受灾点、原有受灾点改变需求信息、原有受灾点撤销应急需求、道路阻塞等动态因素,构建了动态需求条件下的路径优化模型;对已有的求解动态路径优化问题的算法进行分析,结合本文构建的模型做了部分改进,得到了有效的模型求解思路;在算例分析阶段利用改进蚁群算法进行了多种动态事件的模拟分析,验证了算法的有效性。第五部分,实证分析。以某次地震灾害为例,探讨了静态需求与动态需求条件下的路径优化问题;根据实际评价和比较分析,指出了应急物流车辆路径模型和相应算法在解决实际问题中的局限性;结合本文研究,针对当前应急物流管理中存在的问题,从不同的角度提出了相应的对策建议。第六部分,结论与展望。首先,对于本文所取得研究成果进行了总结;其次,分析了本文在研究动态路径优化问题与实证分析中存在的不足;最后,针对物资初始分配、权重优化处理、改进蚁群算法的性能等方面的未来研究方向提出了相关建议。
[Abstract]:The vehicle routing problem is a hot issue in the research field of emergency logistics, emergency rescue emergency plays an important role. The research experts and scholars at home and abroad mainly transit station location and distribution path optimization in two aspects of emergency supplies. Combining with the actual situation, discussed the disaster emergency supplies temporary station location and path integrated optimization problems were constructed in static and dynamic transit demand under the condition of station location and path optimization model, and the improved ant colony algorithm was presented according to the model design, case analysis and empirical research to verify the effectiveness of the model and algorithm. The research results have important theoretical value and the practical significance for effective decision-making and timely response to the emergency department of emergency management. This paper is mainly divided into the following sections: Part of introduction mainly clarifies the background and significance of the research on emergency logistics vehicle routing problem, and summarized the research situation on the basis of the analysis, summarized the current research achievements in this field and further research space, put forward the research methods, research route and the main innovation. The second part. Emergency logistics vehicle routing problem with an overview of the relevant theory. Combined with the status quo of China's research in the field of emergency logistics, emergency logistics involved in research, vehicle routing problem, ant colony algorithm, dynamic programming theory to carry on the summary analysis, to further improve and enrich the theoretical foundation. The third part, the static demand conditions of the emergency logistics center location and routing optimization. Emergency logistics vehicle routing problem of multi supply points with soft time windows multi requirement points and multi material demand. First of all, in emergency For maximum coverage of thought and time, minimizing the cost of thinking, construct integrated station location and multi objective programming model of route optimization of transit emergency supplies; secondly, the multi-objective programming model is constructed, combined with the idea of a variety of intelligent algorithm, the pheromone update and limitation, tabu list sharing, state transition probability has been improved the ant colony algorithm, improve the convergence speed and solution quality of the algorithm, and the corresponding model is suitable for the improved ant colony algorithm is designed; finally, through the example analysis to verify the validity of the model, and through comparative analysis shows the superiority of the improved algorithm. The fourth part logistics path optimization of emergency dynamic demand conditions the introduction of dynamic programming. Based on static path optimization model, the analysis of static and dynamic path optimization problem of path optimization characteristics, this paper proposes to Dynamic path optimization problems; combined with the actual situation, considered a key appear affected by the new and original disaster change demand information, the original disaster emergency revocation requirements, road congestion and other dynamic factors, constructs the path optimization model of dynamic demand conditions; algorithm of path optimization problem of dynamic analysis. This paper constructed the model to do some improvements, got the idea to solve the model effectively; in the example analysis of improved ant colony algorithm is simulated and analyzed several dynamic events using stage, verify the validity of the algorithm. In the fifth part, the empirical analysis of some earthquake disaster as an example, discusses the path optimization problem of static demand with the dynamic demand conditions; according to the actual evaluation and comparative analysis, points out the limitations of the emergency logistics vehicle routing model and its algorithm in solving practical problems; According to the research, in view of the current emergency problems in logistics management, this paper proposes corresponding countermeasures from different angles. The sixth part, conclusion and prospect. Firstly, the research results obtained in this paper are summarized; secondly, this paper analyzes on the study of dynamic path optimization problems and deficiencies in empirical analysis; finally, according to the material initial distribution, weight optimization, improved ant colony algorithm performance and other aspects of the direction of future research and put forward relevant proposals.

【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U116.2;F252

【相似文献】