Nourinejad, MehdiDhaness, Joshua Devindra2025-04-102025-04-102024-09-102025-04-10https://hdl.handle.net/10315/42784Optimal timing for automated vehicle (AV) pilot programs is essential to balance early implementation benefits against the goal of perfected safety. Although AV technology is not accident-free, it shows potential in reducing crash rates. This work introduces a mathematical framework designed to maximize social welfare through AV pilots, integrating elements such as safety metrics, technology diffusion, and learning effects. The framework’s three models address scenarios that social planners may encounter, offering flexibility in balancing initial risks with anticipated safety improvements over time. A case study illustrates these models, emphasizing the framework’s utility in navigating trade-offs like managing higher early failure rates while leveraging learning-based enhancements. By adjusting input parameters, planners can customize pilot designs to meet community needs. Future improvements, such as variable weighting and iterative testing, are proposed to enhance the framework’s adaptability and its capacity to capture critical AV deployment considerations.Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.Electronic Thesis or Dissertation2025-04-10Automated vehiclesAVAVsAutonomous vehiclesAV safetyTransportation planningPilot deploymentSocial welfare optimizationDiffusion modelsLearning effectsDisengagement ratesSocial costsPublic transit integrationRisk minimizationTechnological advancementAV pilot frameworkSocial planner decision-makingAV case studySafety improvementsTrade-offs in AV deploymentModel customizationAV adoptionDiffusionAV pilot design