A framework for acuity-based, individualized patient scheduling

We consider scheduling finite numbers of patients for visits to a medical clinic throughout a workday. Use of individual patient acuity level indicators in the scheduling process is explored to accomodate fluctuations in service requirements. These acuity classifications can be used by schedulers to enable more efficient, individualized service. After analyzing sojourn time behavior for patient populations with arbitrary deterministic arrivals and Erlang service time distributions in a queueing setup, we motivate and develop a Markov model suitable to account for general time-varying service time distributions in the scheduling process. The significance of this work is that it provides a mathematical basis for the analysis of acuity-based scheduling rules currently applied and proven effective in healthcare. This provides a foundation from which to develop and validate new scheduling rules in individualized medicine. The linkage to the proposed Markov model permits the conversion of particular scheduling questions into Markov Decision Problems.