Truck Parking Demand Analysis along Nebraska’s Major Truck Freight Corridor: Utilizing Deep Learning Algorithms on Imagery Data

Abstract This study analyzes the public and private truck parking availability along Interstate 80 (I-80) in Nebraska between 2010 and 2022. The analysis uses a data set containing the National Agriculture Imagery Program (NAIP) and Google Earth imageries of parking facilities located within 1.6 km (1 mi.) of I-80 in Nebraska. Results from the spatiotemporal assessment indicate a pattern of increased truck parking demand along the eastbound locations on I-80, particularly in central and eastern Nebraska. These facilities also experience a higher occupancy rate. These findings suggest that additional truck parking is needed to serve the growing freight movements within and through Nebraska. To facilitate a similar spatiotemporal analysis for other states in the region or for the entire nation, this study trained a computer vision and machine learning model, YOLO (You Only Look Once), to automatically detect trucks parked at a facility given a set of NAIP and/or Google Earth images. The trained YOLO model is publicly available, and it can be used by researchers and practitioners to automate the process of counting trucks parked at a facility. The validation results showed that the YOLO model achieved an accuracy score of 91% and an F1 score of 84.70%, respectively, for parking facilities with up to 22 parked trucks. The findings of this study contribute to an enhanced understanding of the truck parking supply and demand along I-80. This insight is valuable to state transportation and local economic development agencies in developing strategic plans for this major truck freight corridor.