A simulation-based framework for quantifying potential demand loss due to operational constraints in automated mobility services

Automated vehicles are key to unlock a more widespread on-demand service, increasing accessibility also in peripheral areas of large cities. To evaluate how the performance of these services may affect the overall demand in return, multiple dimensions of the transport problem have to be considered. Indeed, despite people may be willing to use Automated Mobility On-Demand (i.e., generating a potential demand for the service), they may be less willing to consistently replace their other travel options if they, for example, experience high waiting times (determined by the performance of the service, i.e., the supply). In this study, we propose a simulation-based framework developed by integrating an activity-based and a dynamic traffic assignment model, designed to frame absorbed and lost demand at a disaggregated level. This allows capturing how the effects of network congestion and fleet constraints may cause a certain portion of the demand to shift to traditional modes of transportation, thus improving, for example, the accuracy of business cases for mobility service design or of hidden patterns of inequality for policymakers and public authorities.