ORCID

0000-0002-9588-6013

Keywords

Ride-hailing Demand Spatial Transferability, Autonomous On-Demand Shuttles, First Mile-Last Mile Connectivity, Passenger-Parcel Integrated Service, Autonomous Shuttle Impact Assessment, Routing Optimization with OR-Tools

Abstract

Prioritizing shared mobility options such as ride-hailing and public transit holds potential to reduce private vehicle dependence and congestion. Ride-hailing effectiveness is often limited by inaccurate spatial-temporal demand predictions and capacity constraints, leading to inefficient fleet use and deadheading miles. Public transit often remains underutilized due to rigid schedules and inefficient accessibility. Advancements in vehicle automation offer potential to overcome these issues and enhance urban mobility. This dissertation explores how autonomous mobility solutions can improve passenger transport and parcel delivery services through four inter-connected studies. The first study investigates cross-city spatial transferability of ride-hailing demand prediction models. The knowledge transferred model, tuned with at least 30% of local data, had higher prediction accuracy, indicating the significance of local tuning. The second study develops a calibration framework for autonomous shuttle simulation using trajectory data and assesses traffic impacts, showing low-speed shuttles significantly affecting mixed traffic with delay time to travel time ratio increase up to 28% and speed reduction up to 10%, which can be mitigated by slightly increasing in shuttle operation speed. The third study develops a dynamic real-time controller for first mile–last mile connectivity with autonomous on-demand shuttles (AODS), integrating mesoscopic simulation-based Dijkstra with greedy- exhaustive insertion heuristics. This framework is more efficient compared to baseline nearest neighbor heuristics in terms of requests served and with a relatively small fleet of 3 shuttles, more than 80% of total requests can be served, maintaining median waiting time below 4 minutes. The fourth study develops a dynamic real-time controller for integrated passenger-parcel service with AODS, where mixed service fleets can significantly increase the service efficiency with 5% higher fleet utilization ratio and 20% less empty vehicle travel ratio than the separate service fleets. This dissertation provides a comprehensive framework to enhance shared mobility efficiency, transit connectivity, and sustainable urban mobility.

Completion Date

2025

Semester

Fall

Committee Chair

Hasan, Samiul

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Department of Civil, Environmental, and Construction Engineering

Format

PDF

Release Date

12-15-2026

Document Type

Dissertation

Campus Location

Orlando (Main) Campus

Subjects

Urban transportation--Research; Transportation--Automation; Urban transportation--Technological innovations; Ridesharing--Evaluation; Paratransit services--Automation

Download

Available for download on Tuesday, December 15, 2026

Share

COinS
 

Accessibility Statement

This item was created or digitized prior to April 24, 2027, or is a reproduction of legacy media created before that date. It is preserved in its original, unmodified state specifically for research, reference, or historical recordkeeping. In accordance with the ADA Title II Final Rule, the University Libraries provides accessible versions of archival materials upon request. To request an accommodation for this item, please submit an accessibility request form.