The Research On Models And Algorithms For Urban Evaluating Basic Land Price And System Implementing Method

The benchmark land price is an important part of the urban land price system. The government periodically announces the benchmark land price, which has the function of guidance and reference to various land prices in the real estate market, having significant meaning to the urban economic development and land market. The benchmark land price is relatively stable. In a certain period of time, the benchmark land price meets the real estate market and urban development. However, with the expansion of urban structure and development of national economy, the benchmark land price bounds to have comparatively big difference from the real price of the land market. Therefore, the benchmark land price must be adjusted periodically.Currently, the measurement and calculation of the benchmark land price in China is independent and is conducted repeatedly. The previous data and results are very difficult for the next use. The task is heavy each time, consuming both time and power. Whether the original benchmark land price accords to the real land market price depends on artificial judgment or compulsory regulations. The price is not adjusted timely. The period for adjustment lasts a few years or even longer. The hysteresis, artificial factors and the related accuracy of traditional benchmark land price has a serious effect on the governmental organizations to objectively and accurately enact the land fees for assignment and other related decisions.There are two methods for land classification and benchmark land price determination, i.e., multi-factors integrated classification and land transaction example classification. Traditionally, the multi-factors integrated classification is usually used. In the process of evaluation, the artificial factors and expert experience would have a big effect on the result. The automation level of evaluation is low. The result cannot reflect the real change of urban land market, and to some extent, limits the daily update of benchmark land price.Based on the computer, network and GIS technologies as well as probability theory, topology, graph theory, statistics and operations research, and combing the urban land use features, the thesis focuses on the study of dynamic evaluation module and algorithm for urban benchmark land price. Use the land transaction example classification method to realize land classification and determination of benchmark land price. Combing the multi-factors integrated classification method, the system canbe applied in practice. The major research content is as follows.1. Study on the basic characteristics of the urban spatial structure;2. Abstracting the urban material space to be the spatial object. Study on the spatial distributed measurement model of these spatial objects;3. Study on the space relationship of the entity factors that constitute the urban space;4. Set up urban spatial system module;5. Study on the urban land use structure features and classification method;6. Study on the modes how urban spatial factors affect the land price. Classify and abstract the spatial factors and establish certain mathematical models for quantification;7. Study on the attenuation model, attenuation and superposing methods of the factor function point value;8. Study on how to deal with the blocking feature (such as railway, river and etc.) in the process of benchmark land price evaluation. Focus on the shortest route algorithm of any point to point or any point to line;9. Study on the models, algorithms and implementation of comprehensive point value simulation model, fixed regional dynamic simulation method and regional gradual fining simulation method;10. Study on the mobile regional interpolation algorithm based on the shortest route. Study and implement the computing flow and related algorithms of market transaction example classification method;11. Study on the measurement for urban land grading benchmark land price. Focus on the benchmark land price measurement methods, such as sample point land price measurement method, sample point land price and land grading measurement method, sample point land price and factor comprehensive point value measurement method;12. Take "Benchmark Land Price Dynamic Monitoring System of Shanghai" as an instance, validating the feasibility of each model and algorithm.The major contributions and points of innovation lie in:1. Factor weighting determination method based on time series;2. Successfully use the shortest route method in the grading price evaluation system, which makes the factor point value and land price analysis morepractical, the result more accurate;3. Make use of the shortest route algorithm to solve the blocking feature issue in land classification and price evaluation;4. Adopt the factor weighting dynamic simulation method, revealing the relationship of land price and factors;5. Use the spatial interpolation algorithm to improve the precision and scientificalnes from points to surfaces;6. Realize the automatic collection of land price factors, improving the automation level of benchmark land price evaluation.The urban benchmark land price dynamic evaluation is an important issue for research. The following issues should be further discussed and studied.1. Technologies and methods for fundamental spatial data dynamic updating;2. How the urban general planning, control planning, detailed planning and governmental decision affect the land price, and how to deal with these issues;3. How to use the expert supporting decision system in the process of classification and price evaluation to support the urban land price analysis and decision making more efficiently.