Industrial Safety

Learn More About the Root Cause and Results of Most Industrial Accidents. Plus IT vs. Control Systems

Share Print Related RSS
This article was printed in CONTROL's July 2009 edition.

More on Industrial Safety

As a former naval nuclear reactor operator, instrument and control engineer at a major petrochemical company and nuclear industry marketing director at a safety instrument vendor, I thought Control’s readers might be interested in the root cause and results of most industrial accidents.

Although several core melts occurred prior to 1979 (ERR-1, Enrico Fermi 1, NRX, Santa Susana Field Lab and others), the accident at TMI-2 was the most serious in the U.S. nuclear power industry. Although no deaths occurred during or following any of these accidents, significant causal reviews were initiated after TMI to preclude another event. Numerous investigations and analyses discovered all of the shortcomings highlighted in Mr. Lipták’s article.

The sweeping changes initiated by the Nuclear Regulatory Commission (NRC) have greatly reduced the probability of repeating another core melt, a fact that significantly diminishes health risk to the workers and public in and around the U.S.’s 104 nuclear power plants. Readers are encouraged to visit the NRC web site for a summary of the TMI accident and all corrective actions implemented since the event (www.nrc.gov/reading-rm/doc-collections/fact-sheets/3mile-isle.html). 

Nuclear regulators in other countries have adopted similar polices. 

It’s unfortunate that others have not learned lessons from these accidents or even those that occurred within their industries. Since the TMI accident 30 years ago, several deadly fossil fuel fires and explosion accidents have occurred at refineries, chemical plants and power boilers (Norco, La., refinery—7 deaths; Auburn, Iowa, metal-plating plant—5 deaths; North Sea oil production platform—167 deaths; Pasadena, Tex., polyethylene plant —23 deaths; Texas City, Tex., refinery—15 deaths; Romeoville, Ill., refinery—19 deaths; and the worst industrial accident in Bhopal, India—over 3,000 deaths.)

Subsequent investigations reveal common causes in all: poor operator training; poorly designed or installed instrumentation; poor maintenance procedures; safety systems bypassed or disabled; and lack of quality control, etc. 

All are testimonies to management’s failure to act. Managers need to understand their responsibility is not just production, performance and profit, but also requires them to actively seek out what might go wrong, and implementing a safety culture and equipment and training that precludes the event and/or mitigates the consequences. The nuclear power industry’s management have successfully learned the lessons of TMI, and implemented excellent policies, procedures, and diverse and redundant equipment to form a defense-in-depth structure and organization, which has gained the confidence of the American public.

J. Troy Martel, P.E.
Safe Operating Systems, Inc.

IT vs. Control Systems

From a comment on the Unfettered blog: There is enough historical evidence to prove that the IT vs. IACS discussion does not yield any fruitful results. Every time I hear a presenter bringing up the topic how IACS is different from IT, I know positively that I am wasting my time. It is the wrong discussion.

We have to face the fact that control system engineers have to carry the burden alone. Let’s focus on helping the Jake Brodskys of this world as much as we can, and try to ignore those self-proclaimed IT security experts who present at hacker conferences as some means of low-budget PR as much as possible.

I am puzzled to see the discussion centering on the electric grid, while the average food plant, automotive supplier or steel company three miles away might be even more vulnerable than some easy-to-protect electric utility.

Ralph Langner
Langner communications, GmbH

Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments