Research On Service Quality Evaluation Method Of NQI Comprehensive Service Information Platform

The National Quality Infrastructure(NQI)comprehensive service information platform provides small and medium-sized enterprises with one-stop public service platform portals by integrating high-quality service resources,including measurement services,standard services,inspection service and conformity assessment.To better the service quality level of the NQI comprehensive service information platform and display the intact platform to users,it is significant to make a scientific and reasonable evaluation of the service quality.Through literature reading,questionnaire survey,model analysis and empirical research,the service quality evaluation of the NQI comprehensive service information platform have been studied in this thesis,which mainly includes the following aspects:(1)Constructing the comprehensive evaluation index system of the NQI comprehensive service information platform.Through reading literature,the selection principles of service quality evaluation indexes of domestic and foreign public service platforms are studied,and the functional characteristics of the NQI comprehensive service information platform are combined.The comprehensive evaluation indexes of service quality of NQI are screened through user demand survey and expert discussion,and the quantitative treatment method of the indexes is given.(2)Building the service quality evaluation model of the NQI comprehensive service information platform.The questionnaire was designed to investigate the user's needs and experience of the service,and experts in related fields of NQI were invited to score the evaluation indicators.The evaluation performance of the fuzzy comprehensive evaluation model and the intelligent evaluation model was compared to determine the evaluation model suitable for the service quality of the NQI comprehensive service information platform.(3)Proposing an intelligent evaluation model of "principal component analysis +improved genetic algorithm + fuzzy neural network(PCA-IGAFNN)".In the traditional Fuzzy Neural Network(FNN)evaluation model used in the evaluation of service quality of NQI comprehensive service information platform,the learning algorithm of the model will not only affect the convergence speed of the model but also lead to the problem of the local optimal solution.In the presented model,the principal component analysis(PCA)is used to remove the correlation both the service quality evaluation indexes of NQI comprehensive service information platform,delete the redundant index factors,reduce the network input,to achieve the purpose of dimension reduction,and effectively overcome the shortcomings of FNN convergence speed.At the same time,the improved genetic algorithm is combined with the FNN.The coefficients of the membership function in the FNN are searched globally by using the adaptive crossover and mutation probabilities,which can overcome the local extremum problem effectively.The simulation in MATLAB and actual evaluation results demonstrated that the simulation test error of the improved PCA-IGAFNN evaluation model has reduced from 0.09386 to 0.00206,which effectively improves the evaluation accuracy.The improved PCA-IGAFNN evaluation model has better evaluation compared with the evaluation model constructed by particle swarm fuzzy neural network and support vector machine.