Background
Solution is about analyzing demographic using the IoT sensor data from offline stores. Customers were looking for a scalable data platform that can be resued and extended for multiple customers and brands. Understanding the user behavior, predicting the visits and oprimizing sales conversions are some of the key use cases.
Solution Implemented
- AWS IoT – Developed streaming pipelines using Lambda and Python.
- S3 based data lake for raw storage.
- Scalable architecture that allows millions of streaming data from several locations.
- Unified model to integrate data from multiple locations.
- AWS RDS for curated data store – data mart.
- Power BI Dashboards with Row level security
- Customizable floor map visualization.
- ML Models for detecting and removing outliers.
- Model for analysing frequently used paths.
Outcomes
Solution helped to percieve clear visibility on the dwell time and head counts on various stores in a mall space. Using the visualization and report, brand owners and mall owners were able to optimize the revenue and opreational staff.
Domain
Retail
Methodology
Agile with sprint every 2 weeks
Duration
1 Year
Resources
6
Project Status
Completed
Technology Used
