Optimizing computed tomography scan scheduling for sustainable energy usage : a simulation study at Medisch Spectrum Twente

This bachelor thesis aims to identify sustainable optimization options for reducing energy consumption of the CT scanners within the radiology department of Medisch Spectrum Twente (MST), a hospital in Enschede, the Netherlands. The research aligns with MST's objectives of sustainability, care quality, education, and scientific advancements. The Dutch government's ambitious targets to reduce CO2 emissions further emphasize the need for energy-efficient healthcare facilities. The study focuses on the scheduling problem associated with the CT scanners and utilizes a Monte Carlo simulation approach implemented in Excel VBA. Two key heuristics, namely the Longest-Expected-Processing-Time-first (LEPT) and Shortest-Expected-Processing-Time-first (SEPT) algorithms, are employed to minimize the total makespan. The simulation generates scan requests and their corresponding processing times, and the results demonstrate that implementing LEPT leads to a 26% improvement in average completion times compared to the current scheduling system. This improvement translates into a potential annual reduction of 16473 kWh in energy consumption, equivalent to the average energy consumption of three three-person households per year in the Netherlands. The research findings highlight the potential of connecting data with the CT scheduling system to improve energy efficiency, reduce CO2 emissions, and create a more sustainable hospital environment. The study contributes to the scientific understanding of scheduling in stochastic environments and provides practical insights for healthcare facilities seeking to optimize energy consumption and scheduling algorithms.