Design And Implementation Of Subway Clearing System

The subway is the most important means of transportation in the big cities.With the formation of urban rail transit networks and the diversification of the investment,the subway's ticket income distribution has been more and more important.Establishing a reasonable,fair ticketing income clearing system will be more close to the reality world,so that the subway operators can get the benefits and encourage the development of the subway.This paper introduces the process of building the subway income clearing system with the platform of Java Web.Design and implement the corresponding modules of site and line management,ticketing data collection,data query and reconciliation.The author participated in the design and implementation of the whole system.focus on:(1)The generating of the effective path:all possible effective paths of each OD(Origin to Destination,starting site to ending site)are obtained through the road network information,using the K shortest path algorithm,and the effective paths also must be the path of no more than three times transfer and the travel time is not more than the shortest path of two times.In the part,the system will use Dijkstra algorithm to get the shortest path of every OD.(2)The machine learning algorithm and cluster of income clearing function:through the above effective path function,the initial cluster number and the initial center point of the clustering algorithm are obtained.Run the clustering algorithm,and compare with the actual passenger flow data,analyze the advantages and disadvantages of the algorithm model,and adjust the parameters.(3)Secret key generation and verification in ticket management:when the ticket card is newly built,it needs to create a secret key and store it in the system.Every time passengers swipe the card,the system need to verify whether the key of the ticket is correct to confirm whether the ticket card is official.The generation of effective paths can get the center node and classification number of clustering algorithm.The clearing module is an important part of the subway ticketing.based on the travel time between the two sites,using the machine learning clustering algorithm,classifying ticket data.The new clearing system is stable in the testing phase,making the ticket income distribution work more convenient.Through the questionnaire to get real data,compared the results of the system with the real data,you can find out the results of clearing improved obviously.