Dynamic Parking Pricing using Transaction Data

Managing on-street parking in dense urban areas poses challenges due to high demand and limited parking space availability, leading to increased congestion and search times for drivers. This thesis explores the efficacy of implementing a dynamic parking pricing policy inside a parking network to mitigate these challenges. Dynamic parking pricing adjusts prices based on parking demands, aiming to balance parking occupancy across different areas. The research investigates the feasibility of utilizing transaction data to predict parking occupancy, eliminating the need for expensive occupancy detection infrastructure. A predictive Neural Network is generated, and a price-setting algorithm is proposed to optimize and change parking prices to increase availability in high-occupied areas and attract drivers to underutilized spaces.