A probabilistic patient scheduling model for reducing the number of no-shows

Abstract

No-shows in medical centres cause under-utilisation of resources and increase waiting times in specialty health care services. Although this problem has been addressed in literature, behavioural issues associated with the patient's socio-demographic characteristics and diagnosis have not been widely studied. In this article, we propose a model that includes such behavioural issues in order to reduce impact of no-shows in medical services. The objective is maximising the health centre's expected revenue by using show-up probabilities estimated for each combination of patient and appointment slot. Additionally, the model considers the requirements imposed by both the health centre's management and the health authorities. An extension of the model allows overbooking in some appointment slots. Experimental results show that the proposed model can reduce the waiting list length by 13%, and to attain an increase of about 5% in revenue, when comparing to a model that assigns patients to the first available slot.

Keywords:

• Appointment scheduling
• no-shows
• overbooking
• health care
• behavioural OR

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Notice that the probability of both patients showing-up is given by $P_{i,j,k}·P_{i,j,k\prime }$, which attains a maximum at $P_{i,j,k}=P_{i,j,k\prime }=\frac{{\pi }_{\mathit{ij}}}{2}$ for any given value of π_ij.

2 We solved the optimisation problems using Cplex 12.7.