Data-Driven Agent-Based Model Development to Support Human-Centric Transit-Oriented Design

Abstract

This paper proposes an agent-based simulation model of activities within an urban environment to evaluate alternative transport-oriented development (TOD) designs and infrastructure investment proposals prepared by urban planners. The students test the model as model users, and the generated model output on the use of the city infrastructure, occupancy of public space, and key data around the pedestrian and vehicle movements can be translated to design modifications by comparing results with desired targets. This provides valuable scenarios to key stakeholders in the design. A particular challenge with using simulation models as part of the decision-making process is the need to include realistic data for the behaviour of the transport system users. To this end, an experiment was conducted in which data on the individual behaviour and activities was collected, which can be integrated into the simulation model to capture realistic responses to TOD proposals. Illustrative results are shown, demonstrating the model can produce meaningful results for planners but also highlights the role of agent-based simulation models in steering the data collection process and engaging with decision-makers.