A deplorable tendency for accident-related risks

Reader feedback tells us that maintaining process-plant safety is an area of perennial concern. That, doubtless, explains why Carolyn Merritt, chair and chief executive officer of the U.S. Chemical Safety and Hazard Investigation Board (CSB; Washington, D.C.), drew a large audience during her keynote address at the Chem Show in New York two months ago. For attendees of that Show who heard her, or who read about her remarks in the Chem Show Daily published by this magazine, some of the commentary below will already be familiar. But it most assuredly deserves the wider audience that this page can give it.

The mission statement of CSB, an independent federal agency that is governed by a Board appointed by the President, is far-reaching: to promote the prevention of industrial chemical accidents that harm employees, damage the environment and endanger the public, through scientific investigations that determine root and contributing cause, and implementation of recommendations to reduce the risk and consequences of accidental chemical releases.

It is true that in the U.S., chemical plants are highly regulated by the federal government and by the states, Merritt acknowledges. Nevertheless, personnel at chemical process companies in many cases exhibit a predisposition for taking accident-related risks, she adds. That tendency is given voice by such sentiments as: “We’ve always done it that way and never had any trouble”, or “That alarm is set too sensitive so I generally ignore it. Never had any trouble”, or “Those ‘pops’ never amounted to anything so we just figured they were normal.”

Proposals for devices to prevent dust explosions are rejected more often than not, she adds, citing testimony during CSB hearings in Washington last June. Dismayingly, some witnesses added that the risk of two explosions in the same equipment in a given plant is unlikely.

Particularly insidious are cases in which a plant almost has a serious incident; for example, a vapor release occurs, but calm wind conditions prevent the material from spreading. Studies cited by CSB show that unless such near-misses are properly regarded as warnings of bad things to come, they can actually encourage risky behavior, because of a natural human desire to see positive probability patterns (the accident was avoided, so it will be in the future, as well) where none exist. But as CSB puts it, Murphy’s Law has not been repealed, and really big accidents are just waiting for the little ones to get out of the way.

There is a related risk, of “normalizing abnormalities”. To avoid it, Merritt notes, it is important to determine the root causes of abnormal occurrences, then to take corrective action, and finally to follow up on the resulting improvements. Readers who wish to pursue this line of thought further are encouraged to read an article, Normalize Deviance at Your Peril, which we published 20 months ago (CE, May 2004, pp. 52–56).

Prevention of accidents depends of hazard recognition, on good engineering and design, on well-thought-out management systems and on human factors – as well as on decisionmaking by upper management plus a good safety climate and culture within a company or plant, emphasizes Merritt. For practical information on how to achieve the prevention goal, one good place to look is CSB’s website, www.csb.gov; many of the recommendations there stem from specific accident investigations.

Nicholas P. Chopey