Skip to main content
Log in

The performance of a generalized Bailey–Welch rule for outpatient appointment scheduling under inpatient and emergency demand

  • Published:
Health Care Management Science Aims and scope Submit manuscript


We are considering the problem of scheduling a given number of outpatients to a medical service facility with two resources servicing outpatients, inpatients, and emergency patients. Each of the three patient classes has associated class-specific arrival processes and cost-figures. The objective is to maximize the total expected reward which is made of revenues for served patients, costs for letting patients wait, and costs for denial of service. For this problem we propose a generalization of the well-known Bailey–Welch rule as well as a neighborhood search heuristic. We analyze the impact of different problem parameters on the total reward and the structure of the derived appointment schedules and address the question of the number of outpatients to be scheduled. The results show that the generalized Bailey–Welch rule performs astonishingly well over a wide range of problem parameters.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others


  1. Bailey NTJ (1952) A study of queues and appointment systems in hospital out- patient departments, with special reference to waiting-times. J R Stat Soc Ser B 14(2):185–199

    Google Scholar 

  2. Cayirli T, Veral E (2003) Oupatient scheduling in health care: a review of literature. Prod Oper Manag 12(4):519–549

    Google Scholar 

  3. Cayirli T, Veral E, Rosen H (2006) Designing appointment scheduling systems for ambulatory care services. Health Care Manag Sci 9(1):47–58

    Article  Google Scholar 

  4. Cayirli T, Veral E, Rosen H (2008) Assesment of patient classification in appointment system design. Prod Oper Manag 17(3):338–353

    Google Scholar 

  5. Conforti D, Guerriero F, Guido R (2007) Optimizing models for radiotherapy patient scheduling. 4OR 6(3):263–278

    Article  Google Scholar 

  6. Cooper WL, Homem-de Mello T (2007) Some decomposition methods for revenue management. Transp Sci 41(3):332–353

    Article  Google Scholar 

  7. Dreyfus SE, Law AM (1977) The art and theory of dynamic programming. Academic, New York

    Google Scholar 

  8. Fries BE, Marathe VP (1981) Determination of optimal variable-sized multiple-block appointment systems. Oper Res 29(2):324–345, March–April

    Article  Google Scholar 

  9. Green LV, Savin W, Wang B (2006) Managing patient service in a diagnostic medical facility. Oper Res 54(1):11–25

    Article  Google Scholar 

  10. Harper PR, Gamilin HM (2003) Reduced outpatient waiting times with improved appointment scheduling: a simulation modelling approach. OR Spectrum 25:207–222

    Article  Google Scholar 

  11. Kaandorop G, Koole G (2007) Optimal outpatient appointment scheduling. Health Care Manage Sci 10:217–229

    Article  Google Scholar 

  12. Kapamara T, Sheibani K, Haas OCL, Reeves CR, Petrovic D (2006) A review of scheduling problems in ratiotherapy. In: Burnham KJ, Haas OCL (eds) Proceedings of the 18th international conference on systems engineering. Coventry University, Coventry, pp 201–207

    Google Scholar 

  13. Klassen KJ, Rohleder TR (1996) Scheduling outpatient appointments in a dynamic environment. J Oper Manag 14:83–101

    Article  Google Scholar 

  14. Klassen KJ, Rohleder TR (2004) Outpatient appointment scheduling with urgent clients in a dynamic, multi-period environment. Int J Serv Ind Manag 15(2):167–186

    Article  Google Scholar 

  15. Kolisch R, Sickinger S (2008) Providing radiology health care services to stochastic demand of different customer classes. OR Spectrum 30(2):375–395

    Article  Google Scholar 

  16. Liao Ch-J, Pegden CD, Rosenshine M (1993) Planning timely arrivals to a stochastic production or service system. IIE Trans 25(5):63–73

    Article  Google Scholar 

  17. Mondschein SV, Weintraub GY (2003) Appointment policies in service operations: a critical analysis of the economic framework. Prod Oper Manag 12(2):266–286

    Google Scholar 

  18. Petrovic S, Leung W, Song X, Sundar S (2006) Algorithms for radiotherapy treatment booking. In: Qu R (ed) Proceedings of 25th workshop of the UK planning and scheduling special interest group. Nottingham, pp 105–112

  19. Rohleder TR, Klassen KJ (2000) Using client-variance information to improve dynamic appointment scheduling performance. Omega Int J Manag Sci 28(3):293–302

    Article  Google Scholar 

  20. Rohleder TR, Klassen KJ (2002) Rolling horizon appointment scheduling: A simulation study. Health Care Manag Sci 5:201–209

    Article  Google Scholar 

  21. Vanden Bosch PM, Dietz DC (2000) Minimizing expected waiting in a medical appointment system. IIE Trans 32(9):841–848

    Article  Google Scholar 

  22. Vanden Bosch PM, Dietz DC (2001) Scheduling and sequencing arrivals to an appointment system. J Service Res 4(1):15–25

    Article  Google Scholar 

  23. Vanden Bosch PM, Dietz DC, Simeoni R (1999) Scheduling customer arrivals to a stochastic service system. Nav Res Logist 46(1):549–559

    Article  Google Scholar 

  24. Welch JD, Bailey NTJ (1952) Appointment systems in hospital outpatient departments. Lancet 259(6718):1105–1108

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Rainer Kolisch.

A Appendix

A Appendix

1.1 A.1 Bounds for the number of neighborhood schedules

We assume N sched  > 2 because otherwise the set of neighborhood schedules would be empty due to Eqs. 12. To specify an upper bound, one has to distinguish two cases:

  1. 1.

    N sched  ≥ N + 2: Let the schedule be such that a 1 ≥ 3 and a i  ≥ 1 for i = 2, ..., N. In this case to construct a neighborhood schedule, every integer a i with i = 1,..., N can be chosen to be reduced by one. Then, to derive a neighborhood schedule, every integer a j with i ≠ j can be chosen to be incremented by one. Applying this procedure, N ·(N − 1) different neighborhood schedules can be found.

  2. 2.

    N sched  < N + 2: Without loss of generality we assume that a 1 = 3 and a i  = 1 for i = 2, ..., N sched  − 2. Every integer a i with i = 1, ..., N sched  − 2 can be chosen to be reduced by one. Note that integers a i with i = N sched  − 1,..., N can not be chosen. Here we have a i  = 0. Again, to derive a neighborhood schedule, every integer a j with i ≠ j can be chosen to be incremented by one. Applying this procedure, (N sched  − 2) ·(N − 1) different neighborhood schedules can be found.

A lower bound on the number of neighborhood schedules can be calculated in the following way. Since we assumed N sched  > 2, there is at least one integer a i which can be chosen to be reduced by one. And again, every integer a j with i ≠ j can be chosen to be incremented by one. Hence, the lower bound on the number of neighborhood schedules is given by N − 1.

1.2 A.2 System utilization

For a given appointment schedule a, we calculate the expected system utilization as follows: First the expected number of scheduled outpatients E(s i ), non–scheduled inpatients E(n i ) and emergency patients E(e i ) arriving in the time interval (t i − 1, t i ] is calculated for i = 1, ..., N. To the sum E(z i ) = E(n i ) + E(s i ) + E(e i ) of the expected number of patients arrived in (t i − 1,t i ] the number of patients not served at time t i − 1 (called the remainder r i − 1) is added. Then the remainder r i for time t i is updated to r i  =  max {0, E(z i ) + r i − 1 − 2}. The expected utilization E(u i ) for slot i is calculated to

$$\begin{array}{lll} E(u_i) & = & \left\{ \begin{array} {l @{\mbox{, if }} l} 1 & r_i > 0\\ \frac{E(z_i)+r_{i-1}}{2} & r_i = 0 \end{array} \right. \end{array} $$

Finally, the expected utilization of the system is calculated to

$$\begin{array}{lll} E(u) & = & \frac{\sum\limits_{i=1}^{N}E(u_i)}{N} \end{array} $$

Table 18 gives an example for the calculation of the expected utilization.

Table 18 Calculation of the expected utilization for N sched  = 8

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sickinger, S., Kolisch, R. The performance of a generalized Bailey–Welch rule for outpatient appointment scheduling under inpatient and emergency demand. Health Care Manag Sci 12, 408–419 (2009).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: