Autonomous on-Demand Shuttles for First Mile-Last Mile Connectivity: Design, Optimization, and Impact Assessment

The First-Mile Last-Mile (FMLM) connectivity is crucial for improving public transit accessibility and efficiency, particularly in sprawling suburban regions where traditional fixed-route transit systems are often inadequate. Autonomous on-Demand Shuttles (AODS) hold a promising option for FMLM connections due to their cost-effectiveness and improved safety features, thereby enhancing user convenience and reducing reliance on personal vehicles. A critical issue in AODS service design is the optimization of travel paths, for which realistic traffic network assignment combined with optimal routing offers a viable solution. In this study, we have designed an AODS controller that integrates a mesoscopic simulation-based dynamic traffic assignment model with a greedy insertion heuristics approach to optimize the travel routes of the shuttles. The controller also considers the charging infrastructure/strategies and the impact of the shuttles on regular traffic flow for routes and fleet-size planning. The controller is implemented in Aimsun traffic simulator considering Lake Nona in Orlando, Florida as a case study. We show that, under the present demand based on 1% of total trips as transit riders, a fleet of 3 autonomous shuttles can serve about 80% of FMLM trip requests on-demand basis with an average waiting time below 4 minutes. Additional power sources have significant effect on service quality as the inactive waiting time for charging would increase the fleet size. We also show that low-speed autonomous shuttles would have negligible impact on regular vehicle flow, making them suitable for suburban areas. These findings have important implications for sustainable urban planning and public transit operations.

💡 Research Summary

The paper tackles the persistent “first‑mile/last‑mile” (FMLM) problem that hampers public‑transit use in sprawling suburban areas. It proposes an autonomous on‑demand shuttle (AODS) service as a cost‑effective, safe, and convenient solution, and develops a comprehensive simulation‑based framework to design, optimize, and evaluate such a service.
The authors integrate a mesoscopic dynamic traffic assignment (DTA) model with a greedy insertion‑heuristic routing algorithm into the Aimsun traffic simulator. The DTA continuously updates link travel times based on real‑time congestion, while the insertion heuristic assigns new passenger requests to existing shuttle tours at the position that adds the smallest incremental waiting time and travel distance. This combination yields near‑optimal routes with far lower computational effort than exact methods.
Shuttle stop locations are generated by clustering residential and commercial parcels using a K‑means algorithm on a NetworkX graph. Fifteen stops are selected as the optimal trade‑off, covering 96 % of parcels within a six‑minute walk. Two transit hubs are also defined to provide the connection to the existing Lynx bus system.
Charging is modeled explicitly. Each shuttle’s state‑of‑charge (SOC) is monitored; when SOC falls below a threshold, the vehicle is dispatched to the nearest charging point. A “charging pool” strategy allows multiple shuttles to share a limited number of chargers, reducing queueing time. The study shows that increasing the number of chargers from one to three cuts average waiting time from 1.8 min to 0.9 min but raises the required fleet size from three to five vehicles, illustrating a clear cost‑service trade‑off.
The case study is Lake Nona, a fast‑growing suburb of Orlando, Florida. Using 2019 mobility data from Teralytics, the authors generate internal and external demand centroids and simulate a 13‑hour period (6 am–7 pm). Under the assumption that 1 % of all trips are potential AODS users, a fleet of three 8‑seat, 15 mph electric shuttles serves about 80 % of FMLM requests, with an average passenger waiting time of less than four minutes and a total travel time of roughly 12.5 minutes per trip.
Impact on regular traffic is assessed by comparing network performance with and without the shuttles. Because the shuttles travel at low speed and occupy a negligible share of lane capacity (≈0.3 % of total), they do not exacerbate congestion; in some corridors a slight speed increase (≈0.5 mph) is observed due to vehicle redistribution.
Key contributions are: (1) a real‑time routing engine that couples dynamic traffic conditions with insertion heuristics, (2) an explicit charging‑strategy model that quantifies the relationship between charger availability, fleet size, and service quality, and (3) an empirical demonstration that low‑speed autonomous shuttles can be introduced into suburban road networks without degrading overall traffic flow.
The findings suggest that a modest pilot deployment (three shuttles) can achieve high service levels, making AODS a viable first‑mile/last‑mile solution. Planners should balance charger infrastructure against fleet size, and low‑speed autonomous shuttles appear especially suitable for suburban contexts where they can improve transit accessibility without compromising existing traffic operations.