Detecting Inaccurate Predictions of Pediatric Surgical Durations

Accurate predictions of surgical case lengths are useful for patient scheduling in hospitals. In pediatric hospitals, this prediction problem is particularly difficult. Predictions are typically provided by highly trained medical staff, but these predictions are not necessarily accurate. We present a novel decision support tool that detects when expert predictions are inaccurate so that these predictions can be re-evaluated. We explore several different algorithms. We provide methodological insights and suggest directions of future work.