21 November 2023 in IEEE Open Journal of Intelligent Transportation Systems
Shared mobility has emerged as a sustainable alternative to both private transportation and traditional public transport, promising to reduce the number of private vehicles on roads while offering users greater flexibility. Today, urban areas are home to a myriad of innovative services, including car-sharing, ride-sharing, and micromobility solutions like moped-sharing, bike-sharing, and e-scooter-sharing. Given the intense competition and the inherent operational complexities of shared mobility systems, providers are increasingly seeking specialized decision-support methodologies to boost operational efficiency. While recent research indicates that advanced machine learning methods can tackle the intricate challenges in shared mobility management decisions, a thorough evaluation of existing research is essential to fully grasp its potential and pinpoint areas needing further exploration. This paper presents a systematic literature review that specifically targets the application of Machine Learning for decision-making in Shared Mobility Systems. Our review underscores that Machine Learning offers methodological solutions to specific management challenges crucial for the effective operation of Shared Mobility Systems. We delve into the methods and datasets employed, spotlight research trends, and pinpoint research gaps. Our findings culminate in a comprehensive framework of Machine Learning techniques designed to bolster managerial decision-making in addressing challenges specific to Shared Mobility across various levels.
6 December 2024 in Transportation Research Part E: Logistics and Transportation Review
18 March 2026 in SSRN Electronic Journal
We study short-run responses to signed transitions between implemented geofenced parking bundles in shared e-scooter systems. Using a within-city causal event design with cumulative local spillover estimates, we analyse exact endpoint replacements such as shifts into no-parking regimes, shifts into no-parking-plus-restricted regimes, and shifts from no parking to allowed parking plus a designated-parking requirement. The analysis uses 34.6 million shared e-scooter trips from two operators and focuses on within-city bundle transitions in Berlin, Duesseldorf, Hamburg, Munich, and Nuremberg. The central result is that the same nominal restriction can either suppress or retain local demand depending on what replaces the removed legal endpoint. Restriction without credible replacement is more likely to contract activity, whereas restriction with credible replacement is more likely to preserve or redirect it. In the confirmatory resolution 8, 14-day, primary, direct-effect family, 10 of 16 direct effects survive BH-FDR adjustment. Three recurring regularities emerge. First, the same nominal tightening reverses sign once bundle content changes. Second, bundles with a restricted component are the most consistently contractionary. Third, restriction-heavy bundles amplify losses in demand hotspots, with high-demand effects up to 3--4 times larger than the all-demand baseline, while replacement bundles scale close to one-to-one. Conceptually, the paper shifts evaluation from coarse policy families to signed endpoint-replacement bundles, helping explain why nominally similar parking restrictions can look heterogeneous across prior studies. Geofenced parking is therefore best understood as an endpoint-replacement design instrument rather than a simple binary restriction.