Simulation Of an Innovative Public Transport System

We report on the use of simulation in the design of in an innovative public transport system. We begin by giving a broad outline of an off-peak public transport system, being designed with the intention of a first implementation replacing weekend services in Canberra, Australia. The proposed system is based on a small number of carefully selected hub nodes within the city. Hub nodes are linked by high-frequency bus service along trunk routes. A multi-hire taxi system is used to shuttle passengers between their origin or destination stop and the nearest hub. In this paper we present the results of a simulation of this system. The simulation was run using actual travel data gathered from weekend bus use in Canberra. The simulation was used to test various design options, and helped to shape several aspects of the overall system design. The main contribution of this paper is to explore the contributions to system design made using insights gathered from the simulation. For example, contrary to expectation, we saw that a small number of nodes gave the best economic performance. We had expected that a larger bus network would allow the use of taxis to be minimised. However, a smaller number of nodes allows the demand to be concentrated better, allowing more effective use of taxis (higher occupancy). Another early design feature that was overturned was the requirement that passengers are always shuttled to their closest hub. In some instances it is more efficient - both for the passenger and the taxi - to travel direct to the passenger's destination. We discuss these and other aspects of the modelling and simulation of the system. We also present preliminary results of the simulations. The results show that for a real demand scenario in Canberra, Australia, the proposed system is able to achieve a one-third reduction in travel time over the current fixed-schedule bus system, without any increase in costs.