Capacity planning with demand uncertainty for outpatient clinics

In this study, we develop a capacity planning model to determine the required number of physicians for an outpatient system with patient reentry. First-visit (FV) patients are assumed to arrive randomly to the system. After their first appointment, the FV patient may require additional appointments, and will then become a re-visit (RV) patient; after each appointment an RV patient will require subsequent visits with a given probability. The system must achieve a set of targets on the appointment lead-times for both FV and RV patients. Furthermore, the system must have sufficient capacity to assure that a given percentile of FV patients is admitted. We develop a deterministic model that finds the required capacity over a finite horizon. We establish the tractability of the deterministic model and show that it provides a reasonable approximation to the stochastic model. We also demonstrate the value from knowing the demands in terms of the required resources. These conclusions are numerically illustrated using real data from the Urology outpatient clinic of the studied hospital.