This project leverages machine learning to enhance delivery time predictions within Brazil's rapidly growing e-commerce sector, aiming to improve customer satisfaction by providing more accurate delivery estimates. Utilizing a comprehensive dataset from Olist, Brazil's largest online marketplace, this study integrates various data types including customer details, order specifics, product information, and geolocation, enhanced with external weather and distance data.
The goal of this project is to develop a predictive model that uses historical order data along with external factors such as weather conditions and public holidays to address the issue of unpredictable delivery times, thereby enhancing customer satisfaction and operational efficiency.
To provide insights into delivery times and their influencing factors (such as weather, holidays, and distance) in Brazil's e-commerce.
Tableau Public Dashboard: https://public.tableau.com/app/profile/libing.gu/viz/OptiDeliverE-commerceDeliveryTimeAnalysis/E-commerceDeliveryTimeAnalysisandPrediction?publish=yes
Please check the file OptiDeliver_Code.ipynb, you can see all the code and the results.
Appreciation to Silvia Xie, Evelyn Li and Mint Kansinee who have invested their time in improving this project.
- Brazilian E-Commerce Public Dataset by Olist: https://www.kaggle.com/datasets/olistbr/brazilian-ecommerce/versions/7
- Brazil Holidays - CalendarLabs: https://www.calendarlabs.com/holidays/brazil/2018#google_vignette
- Weather Data Services: https://www.visualcrossing.com/weather/weather-data-services