Healthcare

In this work, we propose two enhanced ILP-based lean thinking-enabled models, OCPlean1 and OCPlean2, to improve planning efficiency and workload balance in outpatient departments. OCPlean1 incorporates nurses into the model, enabling the delegation of certain examination services from associate senior doctors to nurses. OCPlean2 introduces an adaptive workload balancing strategy aimed at achieving a more equitable distribution of workload after patient transfers. Experimental results demonstrate that OCPlean1 significantly reduces maximum workload under varying conditions, including different numbers of doctors and patients, and fluctuating diagnosis and examination times. On the other hand, OCPlean2 achieves a more balanced workload distribution. These findings contribute to more efficient and sustainable outpatient scheduling models and support better utilization of healthcare personnel.

This work addresses the pressing challenges hospitals face in resource management during the rapid spread of infectious diseases, which increase the risk of cross-infection and financial strain on healthcare systems. To tackle these issues, we introduce a nationwide joint optimization framework of dynamic bed allocation and patient transfer in pandemics, which we refer to as PanOpt. PanOpt presents a novel integer linear programming model for optimizing patient allocation to specific wards based on case severity, dynamic bed allocation across wards, and inter-hospital patient transfers. Our proposed framework minimizes total healthcare system costs while reducing cross-infection risks and ensuring efficient utilization of resources. Our results show that PanOpt outperforms existing techniques, significantly lowering operational costs and providing a robust solution for healthcare resource optimization.