Research On The Decision-making For Urban Land Reserve

Urban land reserve plays a very important role in cultivating and standardizing land market, improving the existent lands of the company, increasing profit of the land for the government, however, high risk exists in urban land reserve system. It is common problem on how to optimize land reserve system, so as to reduce the risk of it in terms of theory and practice. In the past time, the scholars mainly have researched on the necessity,effect or operational mode of the urban land reserve, and lots of achievements have been obtained in the research, but research on the optimization of the urban land reserve system is scarce. To describe the character of deferent part of urban land reserve system, the dissertation will focus on modeling of it. The main work and innovations of the dissertation include:1. BP nerve network is improved based on Accelerating Genetic Algorithm (AGA), and hybrid forecast model is formulated based on improved BP nerve network and ARIMA model, ARIMA model is used to forecast linear character of the data, and improved BP nerve network is used to forecast nonlinear character of the data, and the forecast procedure and steps are also proposed. The hybrid forecast model, which overcomes the limitation of single forecast method, has improved forecast precision.2. On the basis of analysis of urban land reserve problem, problem definition, notation and assumptions are proposed. Decision making model for urban land reserve under limited land reserve funds is formulated according to two land reserve situations based on stochastic inventory theory and methods, solution to the model is also presented. The model can confirm the reorder point for land reserve according to deferent land remise situation, so as to reduce the cost of land reserve and save the land reserve funds.3. On the basis of problem definition, notation, assumptions, and urban land reserve mechanism, the decision making model for confirming the quantity of urban land reserve under the condition of fuzzy income goal and fuzzy cost constraint is proposed based on fuzzy programming theory and methods, also, the solution to the model by using LINDO software is also presented.4. Based on the basic principle of group decision making, induced ordered weighted averaging consistency (C-IOWA) operator is improved based on the theory and methods of fuzzy preference relations, and dual consistency (DC-IOWA) operator is proposed, also, method and procedure of land remise biding evaluation are presented. DC-IOWA aggregation operator overcomes the problem of information lost in the process of aggregating experts'judgment information, so as to improve the efficiency and impartiality for land biding evaluation.