Optimal placement of shared autonomous electric vehicles waypoints using open-source GIS data: a case study of suburban Lille, France

The rise of Shared Autonomous Electric Vehicles (SAEVs) presents a promising solution for efficient and accessible mobility, particularly in data-scarce suburban and rural regions. This paper addresses the critical infrastructure planning challenge of optimally placing static “Waypoints” transportation network nodes where SAEVs pick up and drop off passengers at their nearest origin and destination. We propose a Geographic Information System (GIS)-based methodology that relies exclusively on open-source data to overcome the lack of detailed travel surveys. Building-level points of interest (e.g., shops, schools, public facilities) were extracted from OpenStreetMap to represent potential demand. A weighted p-Median facility location model was then applied to minimize the total passenger walking distance to these Waypoints. A case study in suburban Lille, France, demonstrates that a network of 15 optimally placed Waypoints achieves an effective balance be-tween user accessibility and infrastructure cost. Results underscore the varying importance of different demand generators and show the system’s robustness across weighting scenarios. By integrating open-source GIS data with facility location optimization, this study provides a reproducible, data-light framework for SAEV network planning in regions where conventional data is unavailable.