An MDP model for walk-in patient admission management in primary care clinics

Primary-care clinics with high patient no-show rates and/or high late cancellation rates often adopt overbooking appointments and/or admitting walk-in patients to reduce the negative impact of patient no-shows and late cancellations. In such clinics, a walk-in patient admission policy is needed to determine whether or not to admit a walk-in patient and when a walk-in patient should be seen. In this study, a finite-horizon Markov Decision Process (MDP) model is proposed to optimize the walk-in patient admission policy in such clinics. After examining the properties of the MDP model, optimal walk-in patient admission policies are discovered for several groups of situations, and heuristic admission policies are proposed for the other possible situations. The performance of the heuristic walk-in patient admission policies are compared under different clinic settings. The experimental results demonstrate that admitting all walk-in patients is a simple and good rule in clinics with walk-in patient arrival rates not higher than 20% of service rate. On the other hand, in clinics with walk-in patient arrival rates higher than 20% of service rate, walk-in patients should be admitted when the number of patients waiting for service plus the expected number of punctual arrivals of patients with scheduled appointments is less than the total remaining slots.