AI for Transportation: Implementing Machine Learning from Pilot to Enterprise Scale

This course translates the ten-step roadmap for building machine learning capability at state departments of transportation. Topics include ML fundamentals, use case qualification, organizational readiness assessment, business case development, pilot planning and execution, risk and responsible deployment, and strategies for scaling and sustaining enterprise ML programs. Students will develop the technical management knowledge needed to scope, oversee, and critically evaluate machine learning deployments at a transportation agency.

Learning Objectives

1. Define the principal machine learning paradigms (supervised, unsupervised, and reinforcement learning) and explain the interpretability-performance tradeoff that governs model selection for transportation applications.

2. Identify the two decision gates that govern whether machine learning is the right approach and whether a pilot should proceed, and describe the data, infrastructure, workforce, and funding gaps an agency must assess before committing.

3. Apply the machine learning pipeline steps including data preprocessing, feature engineering, model training, and evaluation, and explain how metrics such as precision, recall, and F1-score inform acceptance thresholds.

4. Analyze the cross-cutting risk factors in responsible deployment including bias, security, privacy, model drift, and liability, and describe strategies for monitoring and mitigating these risks across the system lifecycle.