Truck Parking Occupancy Prediction: XGBoost-LSTM Model Fusion

For haul truck drivers it is becoming increasingly difficult to find appropriate parking at the end of a shift. Proper, legal, and safe overnight spots are crucial for in order them be able comply with Hours Service regulation, reduce fatigue, improve road safety. The lack spaces affects backbone economy because 70% all United States domestic freight shipments (in terms value) transported by trucks. Many research projects provide real-time occupancy information given stop. However, ultimately need know whether will available downstream stop their expected arrival time. We propose machine-learning-based model that capable accurately predicting 30, 60, 90, 120 min ahead. based on fusion Extreme Gradient Boosting (XGBoost) Long Short-Term Memory (LSTM) help feed-forward neural network. Our results show prediction can achieved small errors. Root mean square error metrics 2.1, 2.9, 3.5, 4.1 trucks four different horizons, respectively. unique feature our proposed requires only historic data. Thus, any detection system could also forecasts implementing model.