Multiple Regression Based Dynamic Parking Fare System

This paper explores the development of a web application for dynamic parking pricing, leveraging ReactJS, ExpressJS, MySQL, Google Maps API, and machine learning techniques. The system addresses the challenge of static parking fees by incorporating real-time data and a trained model to predict optimal parking prices. The core functionality hinges on a multiple linear regression model. This model is trained on historical parking data, considering factors such as parking space availability, time of day, base fare, air pollution levels, and user ratings. By analyzing these variables, the model predicts an accurate price for each parking space. The web application utilizes ReactJS to provide a user-centric interface. Users can search for parking spaces and view dynamic pricing in real time. Integration with Google Maps API offers a visual representation of available parking locations along with their corresponding prices. ExpressJS serves as the backend server, facilitating communication between the ReactJS front end and the MySQL database. The database stores historical parking data used for model training, along with real-time updates on parking availability and user ratings. This project contributes to the field of intelligent parking management by proposing a data-driven approach. The system promotes efficient resource allocation and user convenience through dynamic pricing based on various influencing factors. This approach aims to establish a fair and adaptable pricing structure for parking spaces, ultimately improving overall parking management.