This project was built at CxC 2025 @ The University of Waterloo (Continued development post-hackathon). Developed by Nathan Lu.
Scheduling inefficiencies lead to overstaffing, understaffing, and burnout across industries like healthcare and retail. Many businesses struggle to optimize shift coverage without costly manual adjustments. ShiftSync solves this by leveraging multimodal confidence analysis and PostgreSQL-driven analytics to predict peak staffing needs and improve resource allocation.
ShiftSync analyzes historical staffing patterns, demand fluctuations, and external factors to provide dynamic shift recommendations. It stores real-time staffing data in PostgreSQL, enabling seamless Power BI integration for live monitoring and trend visualization.
- Machine learning: Developed a predictive scheduling model in Python to forecast peak staffing needs.
- Database: Used PostgreSQL to store shift data, model outputs, and performance metrics for fast querying.
- Visualization: Integrated Power BI with PostgreSQL to create interactive dashboards with live insights.
Devpost Link: https://devpost.com/software/touchbistro-eda-nathan-lu