Research On Comprehensive Evaluation Of Smart City Development Based On PSF Model

With the rapid development of smart cities in the world,the research of smart city evaluation has become a new hotspot in academia.Overall,there are common problems in construction of smart cities in China,such as lack of awareness,inadequate planning experience and poor comprehensive scheduling.It is of great necessity to establish an evaluation index system and model to assess the comprehensive development of smart city,which is scientific,rigorous,reasonable,effective and able to meet the market demand.Thus,it can clarify the goal of smart city construction,manifest the demands and disadvantages of smart city development,and improve management mechanism to provide guidelines and references for the development of smart city.Aiming at unscientific selection,unified and subjective evaluation measurements and tools for evaluation index of smart city,the t makes the following contributions based on the theory of urban system:First of all,according to the PSF model,the hierarchical structure of smart city is segmented into input support layer,system application layer and core target layer.On the basis of the structure,smart city is divided into six sub-elements,including smart infrastructure,smart management,and so on.Thus,a comprehensive evaluation index system of the smart city development is established.Secondly,based on the characteristics of comprehensive evaluation index system of the smart city development,a judgment matrix is construct using Analytic Hierarchy Process(AHP)and experts' knowledge,which determines the weight of each evaluation index.The comprehensive evaluation index is further synthesized.Then,in order to characterize the non-linear characteristics of the comprehensive evaluation index of smart city development,neural networks are employed to construct the comprehensive evaluation model to quantify the evaluation index more accurately.Finally,AHP,AHP-BP and AHP-ELM are used to evaluate the smart city development among 151 cities in China.The simulation results show that the AHP-ELM model is superior to the other two models in terms of model accuracy and computational cost.