How competition shapes energy efficiency in shared e-scooter systems

Source abstract

Shared e-scooter systems provide a flexible and sustainable solution for short urban trips. However, low operational efficiency often results in long idle periods, during which e-scooters waste energy without providing service. This study investigates how excessive inter-operator competition is associated with such inefficiencies. Using high-resolution operational data from Parma, Italy, where two operators coexisted, we reconstruct full lifecycles and introduce a novel entropy-based competition index. We use an interpretable machine learning framework to assess how competition factors relate to idle energy consumption. The results show that competition-related patterns are associated with higher idle energy consumption, with relocation-based rivalry being most informative for the non-dominant operator. For the weaker operator, over half of battery discharge occurs while idle, with competition accounting for about 70% of the total SHAP importance. These findings call for demand-responsive fleet management and context-sensitive regulation to reduce unnecessary competition and support sustainable micromobility.