Data Driven approach for investigation of Food delivery order cancellations on Zomato platform and development of predictive model to determine order cancellation beforehand.

Abstract

The reasons behind the cancellation of food orders on the Zomato platform in India are examined, based on a dataset of 12,079 food orders placed in 76 cities between January 2023 and October 2024. The study concludes that there is an overall cancellation rate of 32.3%, resulting in revenue loss and inefficiency of operations. Although promotional codes, free delivery offers and price range seem to have an impact, it is not significant in the context of the cancellations that took place. The research is descriptive analysis of several factors that affect the cancellation of orders and appraises the importance of each factor. It also suggests a machine learning based predictive model and a multi-stage analysis process to deal with cancellation issues. The results offer some practical suggestions for platform operators, such as criteria for restaurant engagement, operator capacity building and real-time assessment of cancellation risk when placing an order.