Real time access control of patient service in the pediatrics department

Motivated by patients´ increasing demand and inevitable uncomfortable waiting for the scarce healthcare service, we develop a real-time automatic appointment scheduling (AS) policy considering both the difference and fairness among heterogeneous types of patients. This paper discusses the AS system based on the assumption of independent but non-identical patient arrivials with independent and identically distributed random service durations. The AS policy chooses the patient for the next diagnosis from the waiting pool automatically to minimize the expected total “disutility” according to the current system state. We construct a continuous-time Markov Decision Process (MDP) in a finite horizon to formulate the AS system. Through mathematical analysis the myopic optimal policy and a “switching-curve” are derived in this paper. In the numerical experiment, a case based on the Pediatric clinic of the Peking University Third Hospital (PUTH) is investigated by constructing a simulation model of the AS system. Three heuristic policies along with the myopic policy are testified in both the Markovian settings and the real-world scenario. Our simulation results confirm the effectiveness of the myopic policy that the total “disutility” of patients can be greatly alleviated.