Artificial Intelligence replication of District Hospital Surgical Instrument Sterilisation Load Reassignment Under Unmeasured Biological Indicator Incubation Delays: A Healthcare Safety-Constrained Optimisation Evaluation

District hospital sterilisation units reassign surgical instrument loads between autoclaves when biological indicator incubation results are delayed, but the safety consequences of such reassignment under unmeasured delay are poorly characterised. the present analysis formulates load reassignment as a constrained optimisation problem in which the decision-maker minimises expected instrument unavailability subject to a sterility-assurance constraint that depends on an incubation delay whose distribution is not observed. A replication study was conducted using 600 replications of N = 500 loads drawn from a declared data-generating process with intercept 12.0, exposure 0.42, group -3.1, severity 1.35 and error SD 4.2, and the ordinary least squares estimator was characterised under that process. The estimated bias was 0.01081 for the intercept and 0.00012 for the exposure coefficient; the root mean squared error was 0.85604 and 0.01585 respectively; and empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. The results indicate that the estimator recovers the declared process with negligible bias and near-nominal coverage, which supports the use of the linear safety-constraint surrogate inside the reassignment optimiser. The contribution is a formalisation that separates the sterility-assurance constraint from the incubation-delay uncertainty and shows how a safety-constrained reassignment rule can be evaluated without observing the delay distribution directly.