A Review on Optimizing Order Allocation for Zomato Delivery Partners
Prof. Y.D. Choudhari*1, Vedant Patel*2, Jyoti Rangu*3, Sakshi malve*4, Durgaprasad Sappata *5
*1Professor, Information Technology, K.D.K. College Of Engineering, RTMNU, Nagpur, Maharashtra, India.
*2Information Technology, K.D.K. College Of Engineering, RTMNU, Nagpur, Maharashtra, India.
*3Information Technology, K.D.K. College Of Engineering, RTMNU, Nagpur, Maharashtra, India.
*4Information Technology, K.D.K. College Of Engineering, RTMNU, Nagpur, Maharashtra, India.
*5Information Technology, K.D.K. College Of Engineering, RTMNU, Nagpur, Maharashtra, India.
ABSTRACT
The proposed application aims to enhance food delivery efficiency by optimizing order allocation for Zomato delivery partners through machine learning and real-time analytics. This innovative system leverages historical order data, weather conditions, traffic patterns, and festival schedules to predict high-demand restaurant hotspots, enabling delivery partners to be at the right place at the right time. The model utilizes Random Forest Regressor for demand forecasting, ensuring accurate predictions and improved order distribution. The front-end is developed using Flutter, providing an intuitive and interactive user experience, while the back-end is built using Flask and PostgreSQL, ensuring efficient data handling and scalability. Real-time API integration with Google Maps and Zomato API enhances location-based insights. Experimental results demonstrate a 30% reduction in idle time, a 50% improvement in order allocation efficiency, and a 15% increase in delivery partner earnings, making ZOMASPOT a significant step toward smarter, data-driven food delivery logistics.
Keywords: Machine Learning, Demand Forecasting, Flask, PostgreSQL, Predictive Analytics, Zomato API, Food Delivery Optimization.
