With the advent of Autonomous Vehicles (AV) technology, extensive researches around the design of on-demand mobility systems powered by such vehicles are performed. An important part of these studies consists in the evaluation of the socio-economic impact of such systems for involved stakeholders. A Cost-Benefit Analysis (CBA) methodology has been proposed in previous work and applied in the context of Berlin. In this work, the appraisal methodology is applied to the introduction of AV services in Paris-Saclay, an intercommunity, south of Paris, simulated through MATSim, an agent-based model. In the study at hand, AVs would be implemented as a feeder service, working as a last-mile, last-mile service to public transit, allowing intermodal trips for travellers. To our knowledge, this study is one of the first CBA analyses of an intermodal AV system relying on an agent-based simulation. The introduction of AV in a periurban environment would generate more pressure on the road network but would improve traveller utilities. Some private car users suffer from longer travel times, resulting in a loss in terms of consumer surplus that could offset the new AV users’ benefits, depending on the consumer surplus estimation method. In this configuration, a Stop-Based routing service have a greater ability to benefit from economies of scale than a Door-to-Door routing. Finally, in a periurban environment where on-demand AVs would be added to reduce the access and egress cost of public transit, the social impact would be nuanced for travellers, but externality would increase. This would benefit some travellers but would also involve additional congestion. In that case, a Stop-Based routing seems less impacting than a Door-to-Door one and it could allow to equip a downsized network.