Why Process Safety Demands Unique Indicators

The Fundamental Divergence

For decades, operating companies evaluated site safety through a single lens: personal safety metrics, primarily Lost Time Injury Frequency (LTIF) and Total Recordable Incident Rate (TRIR). This reliance stemmed from the legacy assumption—often derived from Heinrich’s Triangle—that reducing minor workplace incidents proportionally reduces catastrophic events.

In process environments handling hazardous materials, that assumption is invalid. Occupational safety addresses the direct interaction between an individual and their immediate working environment (slips, trips, falls, ergonomic strains, pinch points, and PPE non-compliance). The energy involved is usually localized, individual, and kinetic.

Process safety addresses the operational integrity of systems handling hazardous substances and high-energy inventories (flammable gases, toxic chemicals, high pressure, high enthalpy). It prevents fires, explosions, toxic releases, and major structural failures. These events do not arise from worker carelessness; they stem from the quiet, multi-layered degradation of technical and procedural barriers.

The Historical Turning Point: Baker and Macondo

The 2007 Baker Panel Report investigated BP's North American refining operations following Texas City. It established that corporate leadership conflated personal safety success with process safety health, inadvertently starving process barrier maintenance of capital and scrutiny.

Five years later, on April 20, 2010, the Deepwater Horizon drilling rig exploded, killing 11 workers and causing the Macondo blowout. On the exact morning of the disaster, senior operating company executives visited the rig to present awards celebrating seven years without an OSHA recordable injury. While personal safety protocols were rigorously enforced on deck, critical well barriers—mechanical casing integrity, cement isolation, and blowout preventer annular seals—were impaired.

These disasters cemented industry realization: process safety requires an independent, specialized indicator framework. A zero-injury year tells you nothing about whether your relief valves will lift at setpoint or whether an instrumented high-level trip has been manually defeated.

Diagnostic Traps: What Goes Wrong in Metric Design

When organizations attempt to incorporate process safety, they commonly fall into two traps:

• The Substitution Fallacy: Categorizing high-frequency worker incidents that occur inside a process area as process safety events (e.g., classifying a scald from an open steam hose used for washdown as a process safety containment breach). This dilutes process metrics with occupational noise and hides true process equipment failures.
• The Activity Illusion: Measuring bureaucratic compliance instead of barrier integrity. Tracking 'percentage of operators who attended process safety refresher training' tells you that seats were filled; it does not tell you whether an operator can diagnose an abnormal distillation column level divergence or correctly execute an emergency shutdown procedure.

Process safety indicators must measure either unplanned releases of mass and energy (lagging) or the readiness, availability, and rigor of protective barriers (leading).