设施区位集成计算系统及其应用

发布时间：2018-06-02 02:07

本文选题：设施区位 + 模型与算法体系　；参考：《华东师范大学》2013年硕士论文

【摘要】：设施区位问题是人们日常生产生活中一种常见的问题,这里的设施包括公共服务设施,商业设施,安全防灾设施,医疗设施等。设施区位的优劣能直接影响设施的服务水平和生产效率,像学校,公园,消防站,医院,警察局等公共设施的位置好坏能直接影响服务老百姓的水平,商场,仓库等的选址好坏能够影响企业创造财富的能力,因此正确理解和应用设施区位理论具有很强的现实意义。传统规划人员很少用定量模型解决设施的选址问题,而政府决策者更不是专业人士,让他们用科学的方法解决区位问题显然不切实际。本文的研究目的就是在本实验室以前工作的基础上集成出一个能够包含五大类设施区位问题的计算系统,给决策者提供技术支持。本文主要是在本实验室研究的基础上增加竞争型的设施区位模型,设施区位集成计算系统把设施区位问题分为五类,自由选址问题,布局问题,竞争型问题,邻避型问题,层次型问题。其中自由选址问题包括了单设施重心模型,多设施重心模型和最大福利模型,其算法使用Weiszfeld迭代算法来实现。布局问题分为两类,重心问题与覆盖问题,其中重心问题涉及的模型是P-重心模型,使用节点交换算法求解,覆盖问题包括了P-中心模型,集合覆盖模型,最大覆盖模型,主要是参照Minieka算法求解。竞争型模型也是通过Weiszfeld迭代算法。邻避型的模型分为邻避上限型与邻避下限型模型,其算法参照P-中心模型的算法,只是在求解覆盖问题的时候增加一个D值的约束。层次型的问题包括了集合后备层次型应急设施区位模型与P-中心-重心层次型应急设施区位模型。接着是设施区位集成计算系统的设计,主要是在软件工程思想指导下从需求分析,功能模块设计,数据库设计和核心的公共类设计等几个方面进行详细介绍。其中功能性需求分析主要是参照王铮等(1993)的地理计算平台的需求分析模式,在功能模块设计中主要是介绍输入模块,GIS基本功能模块,模型计算模块,输出模块的设计过程。核心公共类设计主要是启发式算法的基类设计。然后在Visual Studio2008开发平台使用面向对象的编程语言C#与ArcGIS Engine9.3二次开发组件库对系统进行实现。最后将设施区位集成计算系统应用于两个具体的案例,农村中小学选址问题和上海市普陀区大型超市的选址问题。其中农村中小学选址的案例在充分考虑P-重心和P-中心模型优缺点的基础上改进出一个既能保证农村偏远地区又能保证人口相对密集地区学生上学安全的模型,并且将其应用到山东省某镇农村的小学选址。通过对改进后同时具有P-重心P-中心特性的模型与传统的P-重心模型计算结果比较,发现改进后的模型更适合农村地区中小学的选址,它既能够保证农村偏远地区的学生上学距离在一定的约束范围之内,又能保证人口相对密集的镇中心等地区的学生上学方便。偏远地区学生上学通过的道路越短,发生意外的可能性就会降低,学生上学就会更安全,与此同时也能兼顾到所有学生上学的加权距离和相对最小,即整体福利相对最大。
[Abstract]:The facility location problem is a common problem in people's daily production and life. The facilities include public service facilities, commercial facilities, safety disaster prevention facilities, medical facilities and so on. The advantages and disadvantages of the facility location directly affect the service level and production efficiency of the facilities, such as the location of public facilities such as schools, parks, fire stations, hospitals, police stations and so on. The good or bad quality can directly affect the level of the service people, the location of the store and the warehouse can affect the ability of the enterprise to create wealth. Therefore, it is of great practical significance to correctly understand and apply the location theory of facilities. It is obviously unrealistic to let them solve the location problem in a scientific way. The purpose of this paper is to integrate a computing system that can include five major facilities location problems on the basis of the previous work in the laboratory, providing technical support to the decision-makers.
On the basis of the research in the laboratory, this paper mainly adds the competitive facility location model. The facility location integration computing system divides the facilities location problem into five categories, the free location problem, the layout problem, the competitive problem, the adjacent avoidance and the hierarchical problem, among which the free location problem includes the center of gravity model of the single facility and the center of gravity of the multi facilities. The model and the maximum welfare model are implemented with the Weiszfeld iterative algorithm. The layout problem is divided into two categories, the center of gravity problem and the coverage problem. The model of the center of gravity is the P- barycenter model, and the node exchange algorithm is used to solve the problem. The coverage problem includes the P- center model, the set cover model and the maximum cover model, mainly reference M The inieka algorithm is solved. The competitive model is also through the Weiszfeld iterative algorithm. The adjacent avoidance model is divided into the adjacent avoiding upper bound model and the adjacent lower bound model. The algorithm refers to the algorithm of the P- center model, only adding a D value constraint when solving the coverage problem. The hierarchical problem includes the set reserve level emergency facility location. Model and P- center gravity center hierarchical emergency facility location model.
Then, the design of the facility location integration computing system is introduced in detail under the guidance of the software engineering thought, including the requirement analysis, the function module design, the database design and the core public class design. The functional requirement analysis is mainly based on the requirement analysis model of the geographic computing platform of Wang Zheng and so on (1993). In the design of functional modules, we mainly introduce the input module, the basic function module of GIS, the model calculation module and the design process of the output module. The core public class design is mainly based on the base class design of the heuristic algorithm. Then, the two development component library of the object-oriented programming language C# and ArcGIS Engine9.3 is used in the Visual development platform. The system is implemented. Finally, the facilities location integrated computing system is applied to two specific cases, the problem of the selection of rural primary and secondary schools and the location of the large supermarkets in Putuo District, Shanghai.
On the basis of full consideration of the advantages and disadvantages of the P- center of gravity and the P- center model, the case of rural primary and secondary school has improved a model which can guarantee the safety of the students in the remote areas which can guarantee the relatively dense population in the rural areas, and applies it to the primary school location in the rural area of Shandong province. By the improvement, it has the P- weight at the same time. Compared with the traditional P- center of gravity model, the model of heart P- center characteristic is more suitable for the selection of primary and secondary schools in rural areas. It can not only guarantee the distance of students in the remote areas of rural areas within a certain confining range, but also guarantee the convenience of students to go to school in the areas with relatively dense population in the town center. The shorter the path for students to go to school in remote areas, the possibility of accidents will be reduced, and the students will be more safe to go to school. At the same time, the weighted distance and relative minimum of all students can be taken into account, that is, the overall welfare is the largest.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU984;P208

【参考文献】