In shared micromobility networks, such as bike-share and scooter-share networks, using trip data to accurately estimate demand in docked and dockless systems is critical to analyzing how the system is operating, such as identifying the number of dissatisfied users, operational costs, and equity in access, especially for city officials. However, the distribution of available bikes affects the distribution of observed trips. Users may walk from an unobserved cell location to an available bike masking the true location of user demand, and users may look for a bike and not find one, which is unobserved user demand. In collaboration with city planners from Providence, R.I., we present a flexible and interpretable framework to estimate spatial-temporal demand as a spatial non-homogeneous Poisson process that explicitly models how users choose a bike, bridging the gap between the docked and dockless methodology. Further, we present computational experiments highlighting that our method provides more accurate estimates of demand when there is incomplete availability compared to previous methods, and we comment on the results of our algorithm on data from Providence's dockless scooter-share network. Our estimation algorithm is publicly available through an efficient and user-friendly application designed for other city planners and organizations to help inform system planning.

翻译：在共享微出行网络（如共享单车和共享滑板车网络）中，利用行程数据准确估计有桩和无桩系统的需求，对于分析系统运行状况（例如识别不满意的用户数量、运营成本以及访问公平性，尤其对城市管理者而言）至关重要。然而，可用车辆的分布会影响观测行程的分布。用户可能从未观测到的单元位置步行至可用车辆，从而掩盖了用户需求的真实位置；用户也可能寻找车辆但未找到，这部分未被观测到的用户需求。与罗德岛州普罗维登斯的城市规划者合作，我们提出了一种灵活且可解释的框架，将时空需求估计为空间非齐次泊松过程，该过程显式建模用户如何选择车辆，从而弥合有桩与无桩方法论之间的差距。此外，我们通过计算实验证明，当车辆可用性不完整时，我们的方法相比先前方法能提供更准确的需求估计，并讨论了算法在普罗维登斯无桩滑板车共享网络数据上的结果。我们的估计算法通过一个高效且用户友好的应用程序公开提供，旨在帮助其他城市规划者和组织为系统规划提供信息。