Adaptive Hospital Bed Allocation Approach under Multiple Uncertainties

摘要

Due to dynamically changing patterns of bed demand and inpatient length of stay (LOS), hospitals are encountering bed shortage in certain specialties and sub-optimal bed occupancy in others. Bed reallocation is a recurring issue for hospital administrators and it is apparent that a deterministic bed allocation model is not optimal for the future demands of tertiary care and financial sustainability. This work develops a unique two-step decision making framework to optimize hospital bed allocation that can adapt to uncertainty in bed demand and LOS. First, the Erlang loss model (in which the number of beds is fixed and no queuing is allowed) is used to determine the minimum number of beds needed to maintain a desired level of bed availability (accessibility) and bed occupancy (efficiency) for each stochastic scenario. Second, the output from the Erlang model is used as one of the input parameters for a proposed two-stage stochastic mathematical model to determine the allocation of beds (that are optimal over all the uncertain scenarios) by maintaining a balance between cost, accessibility and efficiency. A case study based on a tertiary care hospital in the U.S. is used to demonstrate the effectiveness of the proposed bed allocation approach.