Thesis: Optimal Route Planning for Parking Enforcement Patrol using Reinforcement Learning
Details:
Developed and implemented a Reinforcement Learning framework in Python, utilizing the Multi-Armed Bandit concept to optimize route planning for enforcement vehicles, resulting in response times reduction by 15% and increased resource allocation efficiency by 30%.
Designed and deployed a SQL database in PostgreSQL, integrating real data from Calgary Parking Authority (CPA), streamlining data management for parking lot transactions, observations, and violations. Created API endpoints for seamless data retrieval, using Node.js server with Express.
Built and launched a fully functional web application using React, JavaScript, and TypeScript, serving as a live dashboard for managers to monitor parking lot activities and identify violation trends, enhancing data visualization and insights.
