PAARGAMAN: Passenger Demand Provoked (On-The-Fly) Routing Of Intelligent Public Transport Vehicle with Dynamic Route Updation, Generation, and Suggestion

Abstract

Demand-based public bus service meets the need of passengers with less money, time, and resources by reducing the number of private vehicles on the road. In contrast, dynamic real-time demand-based routing faces challenges like elevated travel time due to the requested assignment based on the paths and vehicle availability. Hence, this research introduces a novel framework named Passenger Influence Bus Service-Intelligent Public Transport System (PIBS-IPTS) for efficient routing of available vehicles based on the demand of passengers. For this, optimal paths are elected from the known routes of the general vehicle through the Cuckoo Search (CS) optimization algorithm. Then efficient route prediction is employed by the Artificial Neural Network (ANN) for passenger flow. Here, the unavailability of the passenger request, such as source location or Destination locations, or the unavailability of both locations is updated while employing the path generation process. The path generation process ensures the reduction of request drops generated by the passenger, which elevates the usage of the general bus service. Here, for the optimal selection of routes from the identified routing paths, a multi-objective function based on traffic density, route condition, and route mobility is employed for the selection of a near-optimal global solution. The method’s performance is analyzed using MAE, RMSE, and MAPE and obtained the best values of 0.69, 0.72, and 0.74, respectively.