Integrated Estimation and Calibration: A Model Formulation Approach

Disaggregated models such as activity-based models play a crucial role in urban planning, infrastructure development, and traffic management. However, these models encounter several challenges, such as sequential development, error propagation, and asynchronicity in data and model components. These challenges can result in inefficiencies and inaccuracies in capturing system dynamics. This paper aims to reconcile these challenges by treating the transport model as a unified entity and considering the interdependencies among parameters. Our methodology advocates for simultaneous estimation of RUM-based components and network parameters, emphasizing calibration to observed data. It involves fitting functions to micro-level data for estimation, ensuring replication of macro-level data patterns. Through concurrent estimation and calibration, our method offers a more comprehensive and reliable framework for transport modelling, bridging the gap between micro and macro perspectives. The results demonstrate that the proposed methodology significantly improves accuracy in predicting travel behaviours and system performance