Shared Automated Electric Vehicles (SAEVs) are poised to revolutionize future transportation. However, potential drawbacks, including increased vehicle usage and the projected shorter vehicle lifespan, introduce critical factors that may impact efficiency and environmental benefits. This research introduces a framework that integrates Agent-Based Modelling (ABM) with Life Cycle Assessment (LCA) for a behaviour-driven SAEV assessment. The ABM simulates regional SAEV operations, informing the LCA of pre- and post-integration scenarios. Sensitivity analysis on fleet sizes, system performance metrics, and Global Warming Potential (GWP) reference values are performed. Findings demonstrate that SAEVs significantly decrease the fleet size and total travel distance by raising the average travel per vehicle. SAEVs integration yields a 75–86% daily GWP reduction without significantly compromising user experience. Over 30 years, fleet replacement needs due to inadequate fleet sizing raised GWP by 170%. Balancing short and long-term environmental impact requires optimizing fleet size to achieve sustainable and efficient service delivery.

The future of road transportation systems faces fundamental changes concerning technological progress and business models. Automated and electric vehicles are coming into the market and evolving towards a service-based mobility system with promises to tackle energy and environmental issues in the mobility sector. Although recent studies have begun to explore the potential impact of shared and privately owned automated and electric vehicles (AEVs) mostly from an operational perspective, little is known about the life cycle impact of such future transport systems. To fill this gap, this paper aims to compare the life cycle environmental impacts of shared vs privately owned AEVs in a regional context. A life cycle assessment (LCA) approach is developed to appraise impact categories with a direct effect on human health, ecosystems, and resources availability. Given that automated vehicles are not yet being used massively, the LCA is applied to synthetic travel demand data to assess the characteristics of privately-owned AEVs and the results of an optimization model that determines the vehicle fleet and driving patterns of shared AEVs serving a regional case-study in the central region of Portugal. Two different vehicle seating capacities - one passenger (non-ridesharing) and four passengers (ridesharing) – are considered to evaluate shared mobility systems. Results show that shared mobility systems yield a potential reduction of up to 42% (with 4 passengers per vehicle) of the system's environmental impacts compared to privately owned automated vehicles. Human toxicity, mineral resource scarcity, and marine and freshwater ecotoxicity are the impact categories with a higher potential of reduction.