All projectsView source
Operations
Inventory Optimisation for Medical Supplies
Hospitals often face either stockouts of critical supplies or excessive inventory holding costs. This project modelled demand patterns to optimise reorder points and quantities.
Projected 22% reduction in holding costs
2025PythonPandasScikit-learnExcel
Dataset
Kaggle: Hospital Supply Chain
Key Questions
- Which supplies have the most variable demand, and how can we forecast them more accurately?
- What are the optimal reorder points and quantities for critical supplies?
- Can we reduce holding costs without increasing the risk of stockouts?
Methods
- ABC-XYZ analysis to classify supplies by value and demand variability
- Time series forecasting with ARIMA and exponential smoothing
- Economic order quantity (EOQ) calculations with safety stock modelling
- Monte Carlo simulation for stockout risk assessment
Results
The optimised reorder model projected a 22% reduction in holding costs while maintaining a 98% service level. High-variability items (XYZ class Z) benefited most from safety stock adjustments.
Demand Variability Index by Supply Category
Recommendations
- Implement dynamic reorder points for the top 50 highest-variability items
- Consolidate orders for low-value, stable-demand items to reduce transaction costs
- Set up automated alerts when usage patterns deviate significantly from forecasts
- Review vendor lead time data quarterly to update safety stock calculations
Limitations
The dataset covers a limited time period, which may not capture long-term seasonal patterns (e.g., flu season demand spikes). Lead time variability from suppliers was assumed constant.