Joint Capacity Planning And Appointment Scheduling Policy With Walk-in Patients

The decision-making process of medical services is a complex and critical process and an attractive field of operational research applications.Medical operations management involves planning and coordinating scarce resources(most of which are expensive and highly specialized),and multiple conflict-of-interest goals need to be considered.In addition,the uncertainty of demands,service time and patients behavior(for example,no show,clinic time preference and priority)in the medical operation management,have introduced additional complexity.Effective appointment scheduling system can effectively alleviate the queue length at the registration,reduce the medical operational costs and relieve the difficult access to medical resources.This problem also attracts the attention of many scholars and medical administrators.In recent years,due to the wide application of open-access scheduling,the joint capacity planning and scheduling problem has become more and more attractive to operational researchers.On the basis of previous studies,this paper addresses the capacity reservation policy for walk-in patients under the assumption of general distributed service time.From the perspective of joint capacity planning and scheduling,the demands of potential walk-in patients are unknown when scheduling the appointment patients.Appointed and walk-in patients are assumed to follow independent arrivals.Patients can be categorized in terms of no-show probabilities,which are easy to obtain from historical data.Based on these data,the hospital reserves capacity for walk-in patients with attending probability over some threshold.Meanwhile,the optimal schedule for appointed patients and reservation policy for walk-in patients are determined by the cost and revenue each patient incurs.During the service process,the patient unserved in the assigned slot overflows to the next slot.Based on the above circumstances,a stochastic programming model is formulated with the objective of profit maximization considering patients waiting costs and doctors overtime costs.This study proposes two improved local searches algorithms,the first one can obtain exact solutions with long computation time,however,algorithm 2 produces good approximation with higher efficiency.Numerical experiments suggest that the proposed algorithm outperforms the commonly used heuristic reservation policy.Sensitivity analysis indicates that lower patients' no-show probabilities,more walk-in patients,lower waiting and overtime cost and more meticulous division of the session bring about higher expected profit of the system.This paper considers the patient's preferences,medical examination and return patients by simulation.Comparisons of patients' service time,waiting time and flow time,the number of arrivals and departures,queue length,resource utilizations and overall capacities shows that the study on joint capacity planning and appointment scheduling problem considering walk-in patients is applicable to the general distributed service time.It is also noted that the overall capacity of the system is similar under all five distributions,and doctors with higher priority will provide treatment services for more patients and their utilization is higher.