This is the authors' abstract. We don't add key points for this paper.
Automated slide stainers usually protect incubation timing by running one technique at a time or processing slides in fixed batches, which leaves the robotic arm idle while samples incubate. This paper presents the dynamic scheduler of a slide-staining instrument with 42 heated reaction chambers served by a single arm. The scheduler ranks the ready tasks with fixed rules and admits a task only if its predicted execution and the incubations it starts do not collide with committed attendances; durations come from a kinematic model of the arm. In 19 runs on seven instruments, covering immunohistochemistry, immunofluorescence, and in situ hybridization, 4526 of 4528 incubations ended within their 60-s tolerance window, and the model predicted run makespans with a mean error of 1.26%. Simulations in a development environment showed that running heterogeneous protocols together took 28% to 66% less time than processing them in protocol batches, and that the required ranking rules gave a shorter makespan than four classical dispatching rules and a random order in 19 of 20 comparisons. At low load, a CP-SAT lower bound lies about 23% below the makespan, mainly because of deliberate holds that synchronize the preparation of reagent mixtures. Each decision took at most 55 ms.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Fernández-Cuevas et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: