Article ID: | iaor201522048 |
Volume: | 45 |
Issue: | 5 |
Start Page Number: | 881 |
End Page Number: | 911 |
Publication Date: | Oct 2014 |
Journal: | Decision Sciences |
Authors: | Klassen Kenneth J, Yoogalingam Reena |
Keywords: | scheduling, stochastic processes |
Appointment policy design is complicated by patients who arrive earlier or later than their scheduled appointment time. This article considers the design of scheduling rules in the presence of patient unpunctuality and how they are impacted by various environmental factors. A simulation optimization framework is used to determine how to improve performance by adjusting the schedule of appointments. Prior studies (that did not include patient unpunctuality) have found that a scheduling policy with relatively consistent appointment interval lengths in the form of a dome or plateau dome rule to perform well in a variety of clinic environments. These rules still perform reasonably well here, but it is shown that a combination of variable‐length intervals and block scheduling are better at mitigating the effects of patient unpunctuality. In addition, performance improves if the use of this policy increases toward the end of the scheduling session. Survey and observational data collected at multiple outpatient clinics are used to add realism to the input parameters and develop practical guidelines for appointment policy decision making.