發(fā)布時(shí)間：2019-05-29 07:33

【摘要】：金融押運(yùn)在金融業(yè)中占據(jù)重要位置,金融押運(yùn)安全是其中的核心問題。目前各家銀行不斷步入金融押運(yùn)社會(huì)化之中,把守押風(fēng)險(xiǎn)轉(zhuǎn)嫁給專業(yè)的保安公司,以降低自身的風(fēng)險(xiǎn)和成本。對(duì)于保安公司,則面臨著押運(yùn)途中、網(wǎng)點(diǎn)交接、庫(kù)房監(jiān)控三個(gè)方面的安全風(fēng)險(xiǎn),在這些風(fēng)險(xiǎn)中押運(yùn)途中以及網(wǎng)點(diǎn)交接是最應(yīng)該注意到的安全風(fēng)險(xiǎn),本文對(duì)此進(jìn)行物流規(guī)劃,從而在金融押運(yùn)中盡可能降低這些危險(xiǎn)。 首先,通過對(duì)金融押運(yùn)物流規(guī)劃的問題分析,建立相應(yīng)的模型。本文以車場(chǎng)、金庫(kù)、銀行網(wǎng)點(diǎn)和金融押運(yùn)車為中心,圍繞這些物體建立他們相應(yīng)的關(guān)系,從而完成金融押運(yùn)物流規(guī)劃的模型建立。 其次,對(duì)于建立的模型,設(shè)計(jì)了分布求解,然后再綜合考慮的設(shè)計(jì)思想。對(duì)于銀行網(wǎng)點(diǎn)過多這一金融押運(yùn)所面對(duì)的特殊問題,采用模糊聚類的方法將距離比較近的銀行網(wǎng)點(diǎn)聚集在一起,這樣對(duì)于銀行網(wǎng)點(diǎn)的分配提供了方便。對(duì)于金融押運(yùn)物流規(guī)劃中最重要的規(guī)劃也是求解過程的主干部分銀行網(wǎng)點(diǎn)分配,采用了模擬退火遺傳算法進(jìn)行求解。這樣提高了算法的收斂速度,而且克服遺傳算法存在易陷入局部極值點(diǎn)等缺陷。對(duì)于求解最少車輛問題,采用了優(yōu)先配合啟發(fā)式方法求解,對(duì)于次優(yōu)配合啟發(fā)式方法雖然計(jì)算速度慢,但是它能夠得到更小的解。對(duì)于金庫(kù)車輛分配給車場(chǎng)的問題,采用基于生成樹的遺傳算法。與基于矩陣的遺傳算法相比,此方法求解此問題較快捷。實(shí)驗(yàn)仿真結(jié)果表明,模擬退火遺傳算法提高了收斂速度,在處理大規(guī)模的問題上有了極大的改善,對(duì)于之前的預(yù)測(cè)得到了很好的驗(yàn)證。 最后,在上述研究成果基礎(chǔ)上,對(duì)金融押運(yùn)物流規(guī)劃子系統(tǒng)的主要技術(shù)與功能等內(nèi)容進(jìn)行了分析。結(jié)合實(shí)際情況,對(duì)系統(tǒng)技術(shù)架構(gòu)、系統(tǒng)功能結(jié)構(gòu)和數(shù)據(jù)庫(kù)進(jìn)行了設(shè)計(jì)。將Java編程實(shí)現(xiàn)的模擬退火遺傳算法、優(yōu)先配合啟發(fā)式方法以及基于生成樹遺傳算法嵌入到相應(yīng)的數(shù)據(jù)分析模塊中,從而使子系統(tǒng)實(shí)現(xiàn)相應(yīng)的金融押運(yùn)物流規(guī)劃的功能。
[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.
【學(xué)位授予單位】：東北大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2012
【分類號(hào)】：F832.33;F252

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 付芳;周泓;;針對(duì)項(xiàng)目式生產(chǎn)調(diào)度的模擬退火算法[J];工業(yè)工程;2010年05期

2 朱南;王龍;;供應(yīng)鏈下中國(guó)銀行業(yè)現(xiàn)金物流管理的模式[J];金融理論與實(shí)踐;2009年04期

3 李書全;孫雪;孫德輝;邊偉朋;;遺傳算法中的交叉算子的述評(píng)[J];計(jì)算機(jī)工程與應(yīng)用;2012年01期

4 許小勇;;混沌模擬退火算法在數(shù)值函數(shù)優(yōu)化中的應(yīng)用[J];計(jì)算機(jī)與數(shù)字工程;2010年03期

5 吉根林;遺傳算法研究綜述[J];計(jì)算機(jī)應(yīng)用與軟件;2004年02期

6 譚吉玉;劉高常;魯美娟;;金融押運(yùn)企業(yè)物流成本控制方法探討[J];科技情報(bào)開發(fā)與經(jīng)濟(jì);2007年26期

7 雷德明;多維實(shí)數(shù)編碼遺傳算法[J];控制與決策;2000年02期

8 劉高常;劉建勝;;從成本角度探討金融押運(yùn)物流的社會(huì)化趨勢(shì)[J];物流科技;2009年11期

9 賀紅梅;保安金融押運(yùn)[J];湖南公安高等�？茖W(xué)校學(xué)報(bào);2000年05期

10 ;Simulated annealing algorithm for detecting graph isomorphism[J];Journal of Systems Engineering and Electronics;2008年05期

，

本文編號(hào)：2487760