Artificial Intelligence replication of District Hospital Neonatal Jaundice Phototherapy Device Scheduling Under Unmeasured Irradiance Degradation: A Healthcare Robustness Evaluation

Phototherapy device scheduling in district hospitals is routinely planned on the assumption that lamp irradiance remains at its rated value, yet irradiance degrades silently between calibrations and the degradation is rarely measured. the present analysis treats the scheduling problem as a robustness question and evaluates it through a replication-based replication of an ordinary least squares estimator applied to a declared data-generating process. Six hundred data sets of 500 observations were drawn from a process with intercept 12.0, exposure coefficient 0.42, group coefficient -3.1, severity coefficient 1.35 and error standard deviation 4.2. The estimator recovered the exposure coefficient with negligible bias (-0.00058) and a root mean squared error of 0.01595, while the intercept showed bias of 0.05172 and root mean squared error of 0.89679; empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. The results indicate that under the stated process the estimator is well calibrated for the exposure coefficient, and that scheduling rules keyed to that coefficient can be evaluated without distortion from estimator noise. Robustness evaluation of phototherapy scheduling should separate estimator behaviour from device-degradation uncertainty, and it specifies the replication design that makes that separation possible.