Order Forecasting System for Vehicles Based on Previous Statistics Requests

The research presents a forecasting demand system for transportation services, based on the neural network usage, a data collection class for gathering data, and a REST API. The data used for forecasting demand for transportation services has been analyzed in detail. Key factors influencing demand dynamics have been identified, including the day time, day of the week, weather conditions, public events, seasonal fluctuations and others. Special attention is paid to the temperature impact and weather conditions on demand changes. The neural network underlying the system is designed to process a large amount of information, both from the internal data storage cluster and from external sources via API. This ensures deep and comprehensive analysis of conditions affecting transportation service demand and allows for accurate forecasts to be generated. The system architecture is considered, highlighting the server key role, which processes user requests and connects to external services and the database. The importance of usage the server for complex calculations and storing large amounts of data is justified, which reduces the load on the mobile device. Significant emphasis is placed on the development of the REST API, which is a key element in ensuring forecast accessibility. This interface allows forecasts to be integrated into various software platforms and applications, making the system universal and flexible. Transportation service dispatchers can use the information obtained to optimize the distribution of transportation vehicles, plan routes, and reduce waiting time. This can lead to increased efficiency of transportation services and reduced overall transportation costs.