Optimization of Soft Mobility Localization with Sustainable Policies and Open Data

A quarter of global greenhouse emissions come from transport, with modern cities producing more than 60% of these emissions. To reduce carbon footprint, several solutions on soft mobility (e.g., optimizing electric vehicles locations) have been proposed using IoT resources and AI techniques. However, these solutions either lack replicability since they ignore city’s needs per area and economic restrictions or lack algorithmic fairness since they account no social criteria (e.g., disabled, age, gender). In this work, we developed AI-based methods to automatically detect the different areas (e.g., rural, urban) and propose two heuristics which incorporate social, environmental and economic criteria of the area in their decision making in the form of sustainability policy templates. Our heuristics solve the p-median problem; they minimize the distance of stations to important points constrained by the cost of new stations. We show that our proposed solution is able to disperse the new stations within the city while covering local neighbourhoods. This work is replicated in two big European cities adapted to different open data and demonstrated by a dedicated visual dashboard.