Supply Chain Optimization with Reinforcement Learning
Care Process & Redesign
Technology
Singapore Healthcare Management Congress
Others
31 December 2021
To optimize stock replenishment using reinforcement learning to balance low stock levels and high availability. AI-driven inventory systems enhance stock control while reducing costs and maintaining service levels.
Year Submitted: 2021
Published Date: 31 December 2021
Tags: Technology, Care Process ＆ Redesign, Digital Health, Data Management, Data Analytics, Digitalisation, Automation, Machine Learning, Operational Management, Supply Chain, Procurement, Inventory Management
About this Content
Aims
To optimize stock replenishment using reinforcement learning to balance low stock levels and high availability.
Background
Traditional SAP MRP processes were inefficient.
Methods
Trained a Deep-Q Network model using 150 SKUs to reduce stock while maintaining availability.
Results
Reduced inventory turnover by 50%, improved stock availability, and optimized warehousing operations.
Conclusion
AI-driven inventory systems enhance stock control while reducing costs and maintaining service levels.
Lessons Learnt
Reinforcement learning is effective in healthcare supply chain management for cost savings and efficiency.
Additional Information
SHM 2021 Shortlisted Project
Keywords
Machine Learning, Inventory Optimization
Innovators' Details
Innovators' Details
Healthcare Cluster(s) | Others |
Organization(s) Involved | Agency of Logistics and Procurement Services |
Platform(s) | Singapore Healthcare Management Congress |
Healthcare Professional Group(s) | Healthcare Administration |
Applicable Specialty or Discipline | Healthcare Administrators |
Project Lead(s) | Ge Zhuo Ran |
Project Member(s) | Wang Min Hao |
Connect with this contributor!
Ge Zhuo Ran - singaporehealthcaremanagement@singhealth.com.sg
Project Attachment
C_562_ALPS_SHM_2021_Supply_Chain_Optimization_with_Reinforcement_Learning.pdf