Evaluating the effectiveness and equity implications of car-use reduction policies through multi-agent simulation

The growing demand for private car transportation in cities has been linked to substantial negative externalities, including emissions, noise, crashes and land consumption. Policymakers have proposed various measures to mitigate these impacts, though only a few were implemented. Meanwhile, the accessibility implications of car-use reduction policies remain largely unexplored in the literature. This study attempts to overcome this gap by presenting a simulation-based framework to evaluate the effectiveness and equity implications of car-use reduction policies. The framework integrates discrete choice modelling, agent-based simulation and disaggregate utility-based accessibility analysis, which enable studying the impact across different population segments. A discrete choice model, based on stated preference data, is used to evaluate mode choices under different policies. The model is embedded within an agent-based simulation, which is calibrated and validated using a national household travel survey. The output is used to evaluate behavioural response and system-wide impacts of the proposed measures, and to quantify accessibility changes for different population segments using a logsum measure. Based on a case study in Munich, Germany, the results indicate that promoting active mobility and reducing speed limits for motorised traffic may be the prevailing strategy, generating modest modal shift and notable environmental outcomes, but causing accessibility losses that vary across population segments. We emphasise that a comprehensive policy analysis should also consider factors like emissions, safety and overall liveability. The study underscores the value of disaggregate analyses for project appraisal, particularly when large-scale implementation is infeasible. Future research should explore preference heterogeneity within the agent-based framework.