Born in the 1960s, recognised in the 1970s, championed for the rest of the century; then the landscape changed, and the abuse started.
ICI, a major UK based chemicals manufacturer, which was a significant industrial force in the latter half of the 20th century, started, in the 1950s, to build and operate facilities which were larger and contained more hazards. Accordingly, when losses of that containment occurred, they did so with increasing impact. Indeed, the fatal accident frequency among workers doubled in the 8 years from 1960. Something had to be done. And it was. And it was called HAZOP.
It worked. Within 12 years, the ICI Fatal Accident Rate had been reduced by over 70% meaning that ICI’s implementation of HAZOP and other supporting techniques saved around 50 future lives.
In 1974 the Flixborough explosion happened, causing 28 deaths, propelling the risks associated with High Hazard Process Industries into the UK public (and therefore government) consciousness. Not only the UK, but indeed the entire industrialised world took notice. HAZOP became, rightly, synonymous with Process Safety. No operator would dare to start up a new HHP facility if at least one HAZOP had been carried out during the design phase. It had become part of the default.
Then, in the early 21st century, the war on Process Hazards, which had been raging for almost 2 centuries, was suddenly won. Incidents levelled off, fatalities flat-lined, incident downtime minimised. The landscape had changed, but the default not. Engineers, such as me, entering the profession in the late 20th century were like the replacement rats in a Skinner Box – the original threat had segued but our response to it had not.
HAZOPs were still a non-negotiable part of any design project schedule (good, of course), but had almost imperceptibly been evolved by the industry stakeholders who were paying attention – the operators.
A typical chronology of several design HAZOPs I have chaired was:
- 1 month out: Aggressive HAZOP schedule produced by Designer
- 1 week out: HAZOP Terms of Reference ‘cut and pasted’ by Designer HSE engineer
- HAZOP starts
- Clients engineers (understandably) major on OPerability aspects, driving for associated recommendations
- Designer impotently bemoans the inexorable schedule slippage
- Most of the Recommendations generated are not safety related
Indeed, metrics from 2 of these projects suggest that the cost of running a 1000 P&ID HAZOP in the UK would be ~$6-8m and that HAZOPs run without the appropriate rigor can overshoot schedule by ~30%.
Can we re-imagine HAZOP for an industrial context where hazards appear to have been tamed? Perhaps….
- Designer to prepare a robust HAZop Terms of Reference, with an assertive emphasis on Safety
- Operability Redirected to earlier P&ID review
- Where credible consequences transgress safety thresholds (injury; fatality, MAH), they should be addressed in a consistent way (commensurate safeguard judgement; LOPA; QRA).
- Other major losses (capital; production; environmental) may optionally be treated similarly.
- Operability issues are captured in a separate non-binding spreadsheet.
HAZop - ‘It’s HAZOP Jim, but not as we know it’
