Abstract

A new integrated demand-responsive transport system is described in this paper, introducing an alternative to overcome the inefficiencies that traditional public transport schemes present in low-density housing areas in and around city centres. On-demand public transport satisfies many of the needs that fixed predetermined routes and schedules cannot fulfill. Both schemas are able to work together, as it is discussed in this paper, being on-demand services a good choice in areas and time intervals where low demand makes fixed routes unsustainable. The proposed solution consists of a flexible integrated management system for road public transport, integrating fixed services, on-demand fixed services and on-demand dynamic services. The architecture of the system and the features of the tailored two-stage algorithm for the dynamic routes are detailed in the paper. The main objective of this algorithm is to solve the multivehicle Dial-a-Ride Problem (DARP) minimizing the number of vehicles and the total distance of the routes. A heterogeneous capacitated fleet is used to serve the customers and the cost matrix is asymmetric, as real-life conditions are assumed. The desired quality of service is set by configurable parameters (e.g., waiting time window and maximum ride time). The performance shown by the algorithm during the tests made, permits offering real-time response to user requests.

BIB_text

@Article {
author = {Harbil Arregui and Estíbaliz Loyo and Oihana Otaegui},
title = {Real Time Software Platform for Dynamic On-Demand Public Transport},
abstract = {
A new integrated demand-responsive transport system is described in this paper, introducing an alternative to overcome the inefficiencies that traditional public transport schemes present in low-density housing areas in and around city centres. On-demand public transport satisfies many of the needs that fixed predetermined routes and schedules cannot fulfill. Both schemas are able to work together, as it is discussed in this paper, being on-demand services a good choice in areas and time intervals where low demand makes fixed routes unsustainable. The proposed solution consists of a flexible integrated management system for road public transport, integrating fixed services, on-demand fixed services and on-demand dynamic services. The architecture of the system and the features of the tailored two-stage algorithm for the dynamic routes are detailed in the paper. The main objective of this algorithm is to solve the multivehicle Dial-a-Ride Problem (DARP) minimizing the number of vehicles and the total distance of the routes. A heterogeneous capacitated fleet is used to serve the customers and the cost matrix is asymmetric, as real-life conditions are assumed. The desired quality of service is set by configurable parameters (e.g., waiting time window and maximum ride time). The performance shown by the algorithm during the tests made, permits offering real-time response to user requests.
}
date = {2011-06-07},
year = {2011},
}