Traffic Jam Detection Based On Fuzzy Analytic Hierarchy Process

With the growth of people’s consumption levels,the number of motor vehicles in the country continues to increase,and traffic congestion has seriously affected people’s normal lives.As one of the key technologies to alleviate urban congestion,traffic jam detection has important theoretical research significance.Because the existing congestion detection model standards are not uniform,and the correlation between indicators is fuzzy,it is difficult to comprehensively apply it to a detection model.Therefore,building a traffic congestion detection model that can integrate various indicators has important practical application value.To solve the above problems,this paper adopts a fuzzy analytic hierarchy process,which builds a decision-making analytic hierarchy model based on the causes of congestion to construct the fuzzy judgment matrix of each indicator,then calculates the weight.Indicators are comprehensively applied to the same detection model.To satisfy the judgment consistency of the index weights as much as possible,the calculated index weights need to be optimized to make the evaluation results of the model more in line with objective facts.However,the traditional eigenvalue iteration method has shortcomings such as missing value precision.Based on the normalization properties of the weight vector,paper converts the correction problem of the weight vector into a constrained optimization problem from the perspective of optimizing the consistency value of the weigh.Constrained optimization problems can be solved quickly,so an improved fuzzy analytic hierarchy process construction method based on whale optimization algorithm is proposed,which realizes the weighted evaluation and weight optimization of the congestion model indicators.The method is verified through experiments.Judgment matrices of different dimensions have good convergence and accuracy performance.Based on the open source real traffic flow data,this paper completes the feature classification of traffic congestion indicators by preprocessing the data,and then uses the method proposed in this paper to calculate the index weights of the congestion detection model,and builds a hierarchical congestion detection model.Finally,using the preprocessed Chengdu traffic flow data,through the comparison and analysis of the congestion index and the trend of traffic indicators,it is verified that the congestion detection model constructed in this paper has better effectiveness and a higher degree of reflection of congestion status.