doc. Ing. Přemysl Šůcha, Ph.D.

There is no doubt that a timely diagnosis is critical to avoid significant harm and death of patients suffering from such diseases as acute ischemic stroke, heart attack, or serious cases of COVID-19 infection. Therefore, medical laboratories must be able to quickly analyze the sample and provide the result. The key performance indicator measuring how fast a laboratory can analyze a sample is called turn-around time (TAT). A factor influencing TAT is how the samples are sequenced in the laboratory workflow. This topic aims to study scheduling problems related to the samples' processing. Our analysis in medical laboratories and review of the literature has shown that this problem is not sufficiently studied while the scheduling of samples' processing can significantly reduce the TAT of high priority samples.

The goal of this thesis is to propose new mathematical models for the scheduling of samples in laboratories, and design efficient scheduling algorithms where the solution space exploration is accelerated by machine learning methods.

• [1] ATLAS Collaboration, "ATLAS Software and Computing HL-LHC Roadmap", https://cds.cern.ch/record/280291