University of Twente Student Theses

Login

Automatically scheduling transplant surgeons at the LUMC while evenly distributing surgical experience

Schuermans, N.P.E. (2018) Automatically scheduling transplant surgeons at the LUMC while evenly distributing surgical experience.

[img]
Preview
PDF
2MB
Abstract:The transplant department of the Leiden University Medical Center (LUMC) experiences difficulties with scheduling the transplant surgeons. This is because the transplant surgeons have more shifts than other departments. Therefore, many surgeries are unexpected and therefore it is not known when these surgeries will take place. The schedules of the different tasks are made per quarter and are currently made by one of the surgeons by hand. The aim of this research is to schedule the transplant surgeons such that the number of surgeries done per year is divided over the surgeons. This is currently not part of their planning environment as equally as possible. To do this, first an integer linear program (ILP) is developed for automatically making the quarter schedules. The number of surgeries performed is not taken into account here. By using this ILP optimal schedules, given the availability of the surgeons, that fulfill all constraints can be made automatically. However, when we look at the fourth quarter of 2018 shortages occur during different tasks in this schedule. Also at the organ removal shift, which is the most important task to schedule. Therefore, the model is adjusted to schedule the organ removal shift by another structure. Then, we do take into account the number of surgeries performed. A stochastic dynamic program (SDP) is developed to do this for a more generic case, where persons must be scheduled over different tasks and during these tasks orders can arrive. The goal here is to equally divide the number of completed orders over the persons at the end of the horizon. We solve this SDP for different small instances with the goal to find a structure in the optimal actions that can be extended to bigger instances. Most of the times the optimal action is such that the expected number of surgeries at the end of the schedule is divided as equally as possible. However, there are some exceptions. This SDP is adjusted to the LUMC case, but we cannot solve this due to the curse of dimensionality. Therefore, we approximate the optimal solution by using the ILP model for scheduling the different tasks over the surgeons and add the condition that we want to divide the expected number of surgeries at the end of the schedule as equally as possible over the surgeons. By using this model, the LUMC can include the number of surgeries performed per surgeon in their planning environment. the LUMC to use the model for scheduling the transplant surgeons and take into account the number of surgeries performed per surgeon. However, to do this a user friendly application should be developed. By doing this optimal schedules, given the availability of the surgeons, can be made that fulfill all constraints, the program is impartial and the surgeons do not have to spend time at scheduling the different tasks over the surgeons.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:31 mathematics
Programme:Applied Mathematics MSc (60348)
Link to this item:http://purl.utwente.nl/essays/76372
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page