There is an increased interest from transit agencies to replace fixed-route transit services with on-demand public transits (ODT). However, it is still unclear when and where such a service is efficient and sustainable. To this end, we provide a comprehensive framework for assessing the sustainability of ODT systems from the perspective of overall efficiency, environmental footprint, and social equity and inclusion. The proposed framework is illustrated by applying it to the Town of Innisfil, Ontario, where an ODT system has been implemented since 2017. It can be concluded that when there is adequate supply and no surge pricing, crowdsourced ODTs are the most cost-effective transit system when the demand is below 3.37 riders/km2/day. With surge pricing applied to crowdsourced ODTs, hybrid systems become the most cost-effective transit solution when demand ranges between 1.18 and 3.37 riders/km2/day. The use of private vehicles is more environmentally sustainable than providing public transit service at all demand levels below 3.37 riders/km2/day. However, the electrification of the public transit fleet along with optimized charging strategies can reduce total yearly GHG emissions by more than 98%. Furthermore, transit systems have similar equity distributions for waiting and in-vehicle travel times.

翻译：交通机构对用按需公共交通（ODT）替代固定路线公交服务的兴趣日益增长。然而，此类服务在何时何地具有效率与可持续性仍不明确。为此，我们提出一个从整体效率、环境足迹、社会公平与包容性角度评估ODT系统可持续性的综合框架。通过将该框架应用于加拿大安大略省Innisfil镇（该镇自2017年起实施ODT系统）进行案例研究，我们得出以下结论：当供给充足且无高峰定价时，在需求密度低于3.37乘客/平方公里/天的情况下，众包ODT是最具成本效益的交通系统；当众包ODT实施高峰定价时，混合系统在需求密度为1.18至3.37乘客/平方公里/天范围内成为最具成本效益的交通方案。在所有需求密度低于3.37乘客/平方公里/天的水平下，私家车出行比提供公共交通服务更具环境可持续性。但公共交通车队电气化结合优化充电策略可减少年温室气体排放总量超过98%。此外，不同交通系统在等待时间与车内出行时间方面具有相似的公平性分布特征。