| The implementation of the last round land use planning has played a significant role in protecting land, ensuring key projects land use, and enhancing awareness of lawful land use, improving the production conditions of land. However, with the rapid development of social and economic, many problems gradually revealed in the last round land use planning, such as the land use planning goal is too simplified, the spatial arrangement of land use is unreasonable, the construction land scale is too small, the gap between the actual demands arid predicted value for land use is very large. So currently a new land-use planning urgently is needed, since the last round land use planning couldn’t meet the requirements of the development of land use situation. In the new land use planning, one of the main tasks is to conduct the deeper research of forecasting the construction land demand, and to provide a basis for land use planning. Therefore, it is necessary to further explore the theory and method of the construction land demand prediction.In this paper, the theory of the construction land demand prediction is explored based on the new land use planning of Bazhong city in Sichuan province. The factors which impact the construction land demand of Bazhong city are selected, and quantitative analysis of the important factors are processed by using the statistical software (Eviews). A multivariate linear regression model and a gray model are built for the construction land demand prediction, the weighted average values of the two models are obtained for the construction land demand of 2010 and 2020 of Bazhong city. The main contents of this paper are as follows:(1) Socioeconomic statistical data from 1998-2009 are selected, considering the natural geographical environment, social and economic development, the land use current status of Bazhong city. The Gross Domestic Product (GDP), investment in fixed assets, the level of urbanization, the total population, the gross industrial values and the Engel coefficient were analyzed in this paper. The results show that the level of urbanization and the Engel coefficient have the most significant impacts on the construction land demands of Bazhong city. The validation results show that R square of the model was 0.992, and the model passed the T-test, which indicates the practicability of this model.(2) The predicted level of urbanization of 2010 and 2020 are 30.18% and 47.17%, and Engel coefficients are 47.74% and 40.98%, respectively. The corresponding construction land demands of 2010 and 2020 are 68760 and 70330 hm2, respectively.(3) The variance ratio (C) and the error probability (p) of the grey model are 0.049 and 1, which indicates the efficiency of the grey model. The predicted values of 2010 and 2020 are 68492 and 70378 hm2, respectively.(4) The construction land demands of Bazhong city of 2010 and 2020 are 68746 and 70290 hm2, respectively. |