Summary

The need for improving efficiency in healthcare is motivated largely by increasing global costs of healthcare. One possibility for improvement is in the optimization of the many schedules found within healthcare. This dissertation focuses on just that for two scheduling problems found within healthcare: the appointment scheduling problem and the master surgery scheduling problem. We first look at the appointment scheduling problem – the problem of assigning time slots to patients booking an appointment at a clinic – examining the various ways in which the randomness of this problem is accounted for, and generalising the problem so that its solutions may be used in a wider range of settings in practice. We consider the application of phase-type distributions as well as simulation and analytical approaches, and we optimize appointment schedules for settings both with multiple healthcare providers, and where patients may arrive in batches rather than one-by-one as is usual. Hereafter, we look at a practical scheduling issue, reporting upon the optimization – via mixed integer linear programming – and subsequent implementation of a surgery schedule for a medium sized hospital in the Netherlands. This problem requires assigning surgical specialties to operate in a given room at a given time during a four-week long repeating schedule; the number of possible combinations of which grows extraordinarily fast, even for a small number of specialties and rooms. In this dissertation, we present the method by which we handled the size of the problem, and pay particular attention to the matter of expectations management throughout the project.