Demand Prediction Based on Historical Data

A model for predicting demand for transportation services based on historical data was developed to plan the rational placement of vehicles in city zones. The model takes into account factors such as seasonality, events affecting demand, weather conditions, and more. By recognizing demand patterns in historical datasets using a sliding window approach, our solution has significantly increased the number of orders for transport services while reducing the time of inactivity for vehicles. As a result, transport companies experienced increased profits.