Truck drivers are essential for keeping U.S. supply chains moving, and the availability of safe and adequate parking is essential for keeping drivers moving. Unfortunately, finding parking remains a persistent challenge; recent estimates find there is only one truck parking space for every 11 drivers. A driver’s inability to locate adequate parking creates real risks; they are forced to park in unsafe areas or pay steep fines for violating their Hours-of-Service (HOS), which are federally mandated and electronically tracked limits on how long they can drive before they have to stop. Current solutions like building new truck parking stops or deploying Truck Parking Information Management Systems (TPIMS) are prohibitively expensive and time consuming to implement. This capstone leverages current truck stop infrastructure to develop a data-driven predictive model that helps drivers determine when and where to find preferable parking. Using historical parking availability and traffic volume data, a driver’s starting and ending locations, preference for amenities, and remaining Hours-of-Service, we calculate a utility score for each stop and rank them from most to least preferable. This gives drivers a clear view of where they should stop based on their personal needs and restrictions. We use these rankings to generate feasibility maps, providing industry stakeholders with powerful visuals that allow them to identify over and under-utilized stops in the current network and determine candidate regions for new truck stop infrastructure. This work demonstrates that a low-cost, simulation-based approach can meaningfully improve parking decision-making and add tangible value to the drivers who keep our supply chains flowing.
