A new model for efficiency evaluation of a bus fleet by window analysis in DEA and data mining
place - asia, place - urban, mode - bus, infrastructure - fleet management, operations - performance, planning - service improvement
Urban transportation, efficiency, data envelopment analysis, data mining, decision tree, case study
Bus fleet performance and efficiency evaluation is one of the issues that have attracted the attention of policy makers and urban managers, who seek to improve the service level for their citizens. In this paper, a hybrid framework is proposed using the Data Envelopment Analysis (DEA) method and data mining techniques to conduct a window analysis for performance evaluation, using the bus fleet in Tehran as a case study. First, the DEA model for the efficiency evaluation of the bus fleet is implemented. To this end, a window analysis is carried out to compare bus fleet performance with the performance of other bus fleets and its own performance for various time periods. The results from the DEA window analysis are then used as the input to the data mining classification method to forecast the efficiency of the bus fleet. Several classification techniques are employed and various methods are used to identify the best algorithm. In this regard, the C5.0 algorithm outperforms the others, and finally the rules hidden in the data set are extracted to forecast the bus fleet efficiency.
Permission to publish the abstract has been given by Taylor&Francis, copyright remains with them.
Alizadeh, S., & Safi, M. (2020). A new model for efficiency evaluation of a bus fleet by window analysis in DEA and data mining. Transportation Planning and Technology, Vol. 43, pp. 62-77.