The business case, on your own numbers
Beyond the SME-weeks a faster, cleaner revalidation gives back, the larger prize is the production that abnormal situations quietly cost you — on the order of $25M a year at a typical refinery (about 3–8% of capacity, most of it avoidable). Set your own numbers below; every input is yours, and the math is shown in full.
Economic Value Calculator
Estimate the annual value at a single facility from one understanding of the safety system — paying off on both sides of process safety: fewer latent gaps that lead to incidents, and fewer self-inflicted trips of correctly-functioning safeguards. Every input is yours to set; defaults reflect published anchors, and the calculation is shown in full.
estimate from throughput
The abnormal-loss pool spans both sides — incidents and operator-induced upsets (spurious activations of correctly-functioning safeguards). Analyst-efficiency savings (~$50–200K/yr) are additive and not counted here. Figures are illustrative and depend on your inputs.
How this is calculated
The model
Per-facility annual value is a transparent product of four factors, each set by you:
Vprod — annual production value at risk (margin or netback at full availability). Labn — share lost to abnormal situations, ~5% (ASM 3–8%; ARC ~5%, of which ~80% avoidable). faddressable — share of the abnormal-loss pool addressable by hazard intelligence, after ceding violations, slips, fatigue, and equipment/alarm failures to other interventions. c — realized capture, calibrated by pilot.
faddressable is derived from a disaggregation of incident causation: ~80% of process incidents involve human failure as a primary cause; removing the error types outside hazard intelligence's reach and retaining only decisions mediated by a missing or un-maintained documented hazard basis leaves a conservative count-weighted residual of ~9%. The U.S. Chemical Safety Board corpus shows that failure mode contributing to roughly half of major incidents; because losses concentrate in that catastrophe tail, the loss-weighted figure is 20–30%. We default to the 9% floor.
Sources: ASM Consortium and ARC Advisory (abnormal-situation losses); AFPM, Kimberlite, SBM Offshore, EIA, ICIS (facility economics); peer-reviewed human-factors literature, UK HSE, SPAR-H / PetroHRA, HFACS-OGI (causation); Kaszniak 2010, Baybutt 2016, Amyotte et al. 2011, El-Halwagi et al., and CSB investigation reports (incident-record cross-check).
Modern safety systems work. But the complexity they create is drowning you.
Modern process safety works. More protection, more instrumentation, more rigorous analysis, more disciplined change management — it has driven incidents down.
But it came with a hidden cost. Between revalidations, every change, incident, and revision pulls the hazard basis a little further from reality — and keeping it current, consistent, and defensible is a relentless, under-resourced grind.
Quietly, between revalidations
Inconsistent safeguard credits, un-reconciled MOCs, and stale scenarios accumulate where no one has the hours to look — until a revalidation or an audit forces the reckoning.
Nuisance trips, not safety events
When that context doesn’t reach the operator, a correctly-functioning safeguard trips the process exactly as designed — a spurious trip that takes production with it, with no safety event at all.
Abnormal situations of both kinds run on the order of 3–8% of capacity — tens of millions a year at a typical asset, most of it avoidable
The trade-off, dissolved
In functional-safety terms you already manage this trade-off: more protection raises the SIL, but it also raises the spurious-trip rate — every lever moves you along the same curve (IEC 61511). One understanding of the safety system improves both at once.
We don’t ask you to trade safety for uptime — and we never strip protection. We retire the contradictory or redundant safeguard credits that add nuisance trips without adding protection, and we shift the curve.