Student: Pauline Hoogeland
School: Technical University Eindhoven (TUE), The Netherlands

To complete my master 'Operation Management and Logistics' at the Technical University Eindhoven I wrote a master thesis on the work I performed at the radiotherapy department of the Catharina Hospital in Eindhoven.

The purpose of the study was to improve the waiting time performance at the department. The waiting time studied is the time between referral of the patient to the department and the first radiation treatment. Between these two events several activities are required. First an intake meeting is planned with a radiation oncologist. Second, scans of the patient are made to determine the part that needs radiation treatment. In the final step a plan is made of the radiation treatments, in which the doses and the total number of treatments is determined.

There are different types of patients. Priority 1 patients require acute care, priority 2 patients require sub-acute care and the other patients are classified as a priority 3 patient. A maximum waiting time norm is assigned to each type of patient by the Dutch Society of Radiotherapy and Oncology (NVRO). Besides the three different urgencies, a distinction is made between different types of cancer. The four largest tumor groups are prostate, breast, lung and rectum.

All patients that received radiation treatment in 2005-2007 were registered by the department (data such as tumor group, dates of activities and urgency of the patient). Demand (number and type of patients being referred per day) could therefore be analyzed relatively easily. After analysis it became clear that the number of referrals per day differed significantly between the types of tumor groups.

Also analysis was done of the supply side, such as the availability of radiation oncologists (needed for intake meetings), the CT scan and the linear accelerators (linacs) needed for the radiation treatment itself. The linac delivers a uniform dose of high-energy x-ray. These x-rays can destroy the cancer cells while sparing the surrounding normal tissue.

In the analysis a discrepancy was found between the demand and the distribution of available intake slots per day. Also availability of intake slots varied during a year, due to holidays. From interviews it became clear that radiation treatments for urgent patients were harder to plan, since they are not known long in advance. Therefore two redesign questions were formulated:

1. How can a different distribution of intake slots per day of the week make a difference in the waiting time performance?
2. Does reserving capacity for high urgent patients on the linacs make a difference in the waiting time performance of patients?

These two questions were answered by simulating the processes in Enterprise Dynamics. Two models were made, one to simulate the intake meetings, and another one for the radiation treatments on the linac.

For the simulation models, data was required to use as input variables. Based on the analysis, numbers for demand (per tumor group) are generated according to the Poisson distribution. First the current situation was simulated to see whether differences in capacity planning would result in an improvement. Distributing the intake slots over the week according to the expectation of demand, simulation showed a significantly improvement of the time until intake. Also simulation showed that when reserving capacity on the linacs for urgent patients, their waiting time also decreased.

Simulation was very useful in this situation since the effects of changes in the planning clearly showed whether an improvement was found or not. The radiotherapy department was very happy to see that improvements were possible and plans to implement the recommendations made.

Pauline Hoogeland MSc.