Research On The Price Evaluation Method Of Used Truck Considering Transaction Big Data

The price evaluation method of used truck is the core link of the technical specification of used truck appraisal and evaluation.Currently,the price evaluation of used truck in China mainly relies on traditional asset evaluation methods such as the current market price method and the replacement cost method.There are few effective methods for evaluating the use performance of used truck,and there is no price evaluation model for used truck considering transaction big data.Therefore,it is necessary and urgent to develop a price evaluation method for used truck.Based on the existing technical specification of used truck appraisal and evaluation,sorted out the appraisal and evaluation process,and analyzed the application characteristics of the current market price method and replacement cost method in price evaluation activities.From four perspectives: policy factors such as emission indicators,social factors such as the freight market and e-commerce development,vehicle factors such as accumulated mileage,accumulated driving time,and market factors such as vehicle ownership and brand,this paper analyzed the logical relationship between various factors and changes in the price of used truck,and constructs a price evaluation index system for used truck.Introduced BP neural network to establish a price evaluation model for used truck,analyzed the advantages and disadvantages of BP neural network model in predicting the price of used truck,used genetic algorithms to optimize the BP neural network,established a GA-BP neural network model,and demonstrated the feasibility of the 2 models in evaluating the price of used truck.Selected 3 typical ecommerce trading platforms in China,used Python and data collectors to capture and conducted SQL Server data cleaning,screening,and rationality judgment to obtain 9016 big data on used truck transactions,and developed quantitative standards for the value evaluation index system to achieve data quantification.Taked transaction big data as an example,used MATLAB software to train and test the BP neural network model and the GA-BP neural network model,compared and analyzed the mean square error,correlation coefficient,and program run time of the evaluation results,verified the practical applicability of the 2 price evaluation models,and used MIV to analyze the correlation between different indicators and the evaluation results of used truck price.The research results show that the mean square errors of BP neural network model and GA-BP neural network model in predicting the price of used truck are 0.0086133 and0.0021664,respectively,reducing by 74.85%.The total correlation coefficients are 0.95219 and0.95816,respectively,increasing by 0.63%.The prediction accuracy of the optimized model has significantly improved,making it more suitable for the price evaluation business of used truck.The GA-BP neural network model can also provide reference for the price evaluation of trucks in the fields of state-owned asset disposal,transportation management,vehicle replacement,judicial auction,insurance claim valuation,and other fields.