金融押运物流规划问题的研究与应用

发布时间：2019-05-29 07:33

【摘要】：金融押运在金融业中占据重要位置,金融押运安全是其中的核心问题。目前各家银行不断步入金融押运社会化之中,把守押风险转嫁给专业的保安公司,以降低自身的风险和成本。对于保安公司,则面临着押运途中、网点交接、库房监控三个方面的安全风险,在这些风险中押运途中以及网点交接是最应该注意到的安全风险,本文对此进行物流规划,从而在金融押运中尽可能降低这些危险。首先,通过对金融押运物流规划的问题分析,建立相应的模型。本文以车场、金库、银行网点和金融押运车为中心,围绕这些物体建立他们相应的关系,从而完成金融押运物流规划的模型建立。其次,对于建立的模型,设计了分布求解,然后再综合考虑的设计思想。对于银行网点过多这一金融押运所面对的特殊问题,采用模糊聚类的方法将距离比较近的银行网点聚集在一起,这样对于银行网点的分配提供了方便。对于金融押运物流规划中最重要的规划也是求解过程的主干部分银行网点分配,采用了模拟退火遗传算法进行求解。这样提高了算法的收敛速度,而且克服遗传算法存在易陷入局部极值点等缺陷。对于求解最少车辆问题,采用了优先配合启发式方法求解,对于次优配合启发式方法虽然计算速度慢,但是它能够得到更小的解。对于金库车辆分配给车场的问题,采用基于生成树的遗传算法。与基于矩阵的遗传算法相比,此方法求解此问题较快捷。实验仿真结果表明,模拟退火遗传算法提高了收敛速度,在处理大规模的问题上有了极大的改善,对于之前的预测得到了很好的验证。最后,在上述研究成果基础上,对金融押运物流规划子系统的主要技术与功能等内容进行了分析。结合实际情况,对系统技术架构、系统功能结构和数据库进行了设计。将Java编程实现的模拟退火遗传算法、优先配合启发式方法以及基于生成树遗传算法嵌入到相应的数据分析模块中,从而使子系统实现相应的金融押运物流规划的功能。
[Abstract]:Financial escort occupies an important position in the financial industry, and the security of financial escort is one of the core issues. At present, banks continue to enter the socialization of financial custody, transfer the risk of custody to professional security companies, in order to reduce their own risks and costs. For the security company, it is faced with the security risks in three aspects: on the way of transportation, the handover of the network and the monitoring of the warehouse. Among these risks, the way of escorting and the handover of the network is the security risk that should be paid attention to most. This paper carries on the logistics planning to this. In order to reduce these risks as much as possible in financial transport. First of all, through the analysis of the problems of financial escort logistics planning, the corresponding model is established. In this paper, the car yard, treasury, bank network and financial escort vehicle as the center, around these objects to establish their corresponding relationship, so as to complete the model of financial escort logistics planning. Secondly, for the established model, the distributed solution is designed, and then the design idea is considered comprehensively. For the special problem of excessive bank outlets, fuzzy clustering method is used to gather the bank outlets close to each other, which provides convenience for the distribution of bank outlets. The most important planning in the planning of financial transportation logistics is also the distribution of bank outlets in the main part of the solving process, and the simulated annealing genetic algorithm is used to solve the problem. In this way, the convergence speed of the algorithm is improved, and the defects of genetic algorithm, such as easy to fall into local extreme point and so on, are overcome. For solving the minimum vehicle problem, the priority cooperation heuristic method is used to solve the problem. For the suboptimal cooperation heuristic method, although the calculation speed is slow, it can get a smaller solution. The genetic algorithm based on generative tree is used to solve the problem of the allocation of treasury vehicles to the yard. Compared with matrix-based genetic algorithm, this method is faster to solve this problem. The experimental results show that the simulated annealing genetic algorithm improves the convergence speed and greatly improves the processing of large-scale problems, which is verified by the previous prediction. Finally, on the basis of the above research results, the main technology and functions of the financial escort logistics planning subsystem are analyzed. Combined with the actual situation, the system technical architecture, system function structure and database are designed. The simulated annealing genetic algorithm implemented by Java programming, the heuristic method and the genetic algorithm based on generative tree are embedded into the corresponding data analysis module, so that the subsystem can realize the function of the corresponding financial escort logistics planning.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.33;F252

【参考文献】