Batch risk‑based inspection scheduling struggles to capture a critical failure class: facilities that pass recent inspections but later develop severe violations. Traditional systems recompute risk on fixed planning cycles using static historical data, leaving blind spots between cycles. To address this, we introduce the Continuous Inspection Risk (CIR) system, an event‑driven scheduling pipeline that dynamically re‑ranks the inspection queue as new data arrives. CIR combines a decay‑weighted historical score with a stream of real‑time consumer reviews triaged by a language‑model classifier. A key structural contribution is the temporal partition rule, which strictly excludes reviews predating the most recent inspection to prevent outdated signals from skewing current risk. We evaluate CIR on 9,505 routine inspections in Santa Clara County, California (2024‑2026). Using history‑only scoring under a 30% inspection budget reduces the mean time‑to‑detection of major violations by 69 days (20.9% reduction) but fails to anticipate severe vermin‑related closures (0/8 detected). Incorporating post‑inspection review signals predicts 7 of 8 vermin closures with an average lead time of 117 days, firing on the exact day of review publication in four cases. Adding a review corroboration rule improves precision from 0.70 to 0.78 without loss of recall. We also analyze critical implementation challenges, documenting a data‑hygiene pitfall (stale review leakage across inspection boundaries) and an evaluation limitation (unobserved facility self‑remediation). This work provides the first case‑level evaluation of event‑driven inspection triggering using open data and offers a reproducible benchmark for review‑augmented public health systems.
Review: CIR demonstrates the promise of fusing real‑time social signals with decay‑based historical models, delivering a more agile risk perception capability for public health inspections.