Research On Procurement Mode Selection Of Old City Reconstruction Project Based On BP Neural Network

With the advancement of urbanization,problems such as high population density,poor housing quality and outdated infrastructure have gradually emerged in the old urban areas,which makes the development gap between the old and the new urban areas more and more large.On the other hand,with the gradual expansion of the city scale,land resources are becoming increasingly scarce,prompting the transformation of the old city has become an inevitable trend of urban development.In order to narrow the development gap between the old city and the new city,improve the utilization rate of land resources,so that more people can enjoy a comfortable and comfortable urban life,the transformation of the old city is steadily advancing,which has become a long-term work in China’s urban development.This paper is committed to the research on the procurement mode selection method of old urban renewal projects,and provides decision-making basis for the implementation of old urban renewal projects.Based on the domestic and foreign research results about the old city reconstruction project procurement mode selection,this paper expounds the theoretical basis of the old city reconstruction and procurement mode selection;according to the project procurement mode,five commonly used modes of the old city reconstruction are summarized,including DB mode,DBB mode,EPC mode,PPP mode and PMC mode,which provide theoretical support for determining the output layer of the procurement mode selection model of the old city reconstruction project.The common methods of index screening and mode selection are compared and analyzed.The principal component analysis method is determined as the method of index screening,and the BP neural network is determined as the method of mode selection.In order to construct the index system of procurement mode selection for old urban renewal projects,this paper preliminarily selects 45 indicators related to the selection of transformation mode through literature research.45 three-level indicators were further screened,questionnaires were issued,and experts were invited to score the correlation between the indicators and the transformation mode.SPSS 24.0 software was used to conduct principal component analysis of the primary indicators from five aspects of external factors of the project and the owner’s own factors to form the final index system.The index system constructed in this study can objectively reflect the relationship between each index and the selection of transformation mode,and can provide a basis for determining the input layer of the procurement mode selection model of the old urban reconstruction project.In the part of model construction,this study uses MATLAB R2020b software to determine the network layers of the input layer,the hidden layer and the output layer,select the network learning parameters,and select the transfer function to construct the selection model of transformation mode according to the procurement mode selection index system of the old urban renewal project.In the empirical analysis part,this study selected272successful old city reconstruction projects in Jiangsu Province as a typical case,with the help of MATLAB software for the case to select the appropriate transformation mode,and verify the feasibility of the model,put forward reasonable suggestions for the implementation of the model.The test results show that the test accuracy of the procurement mode selection model of the old city reconstruction project constructed in this study is good,the output results can basically meet the accuracy requirements,the error is low,and the overall accuracy is 0.98555.It shows that the model is scientific and reasonable,and has reference value,which can provide decision-making basis for the procurement mode selection of the old city reconstruction project in the future.