PhD Defence Eline Tsai | Improving clinical chemistry laboratory logistics

Improving clinical chemistry laboratory logistics

_{Eline Tsai is a PhD student in the Department of Mathematics of Operations Research. (Co)Promotors are prof.dr. R.J. Boucherie from the Faculty of Electrical Engineering, Mathematics and Computer Science, prof.dr. Y.B. de Rijke and dr. A.N. Tintu from Erasmus Medical Center and dr. D. Demirtas from the Faculty of Behavioural, Management and Social Sciences.}

Clinical chemistry is a multidisciplinary field encompassing several sub-disciplines such as immunology, hematology, and clinical biochemistry. Laboratories perform diagnostic tests on body fluids like blood and urine, playing a critical role in diagnosing, monitoring, and predicting diseases. High-quality testing and timely reporting are essential. Laboratory professionals support clinicians in selecting appropriate tests and interpreting results, meaning laboratory performance directly impacts broader medical care.

As global healthcare demand rises while financial and human resources remain limited, there is increasing pressure to deliver high-quality testing rapidly and cost-effectively. Laboratory design is context-dependent, varying with sample volume, sample type, test requirements, and physical constraints. Given their central role in healthcare, optimizing laboratory operations is essential.

This thesis focuses on optimizing logistics in clinical chemistry laboratories using Operations Research techniques. First, relevant performance indicators (PIs) are identified, incorporating insights from both medical and production process literature. Second, methods are developed to measure these PIs and identify improvement opportunities using process mining. Turnaround time (TAT) emerges as the most critical PI and becomes the main focus of subsequent analyses. Third, the impact of different PIs on TAT is quantified, enabling accurate predictions that support proactive interventions and more efficient use of scarce resources. Fourth, sample assignment to analyzer lines is optimized to maximize the proportion of samples meeting predefined TAT target, addressing the problem of assigning jobs to parallel servers under information delay. Finally, optimal static sample routing within analyzer lines is studied to minimize mean TAT, with results applicable to broader laboratory design.