Reader feedback: Safety requires inspections

Nov. 9, 2015
A reader tells us what was missing from one of our latest articles.

As a chemical engineer with extensive chemical processing industry (CPI) experience in operations management and process safety, I read "Beyond Bhopal" [Aug. '15, p. 32] with interest. Bill Mostia covered a range of important facets of process safety, and made incisive observations in certain areas, but I would like to add two comments.
First, it would have been helpful for the two major incident tables to have listed the underlying cause(s) of each safety event. In the case of the Bhopal tragedy in India, the underlying cause was internal sabotage in which water was introduced into the methyl isocyanate tank using a water hose. It was not a process incident in the classic sense.

Second, the author failed to list the all-important, comprehensive field inspections of process piping and equipment for indications of incipient changes, which could eventually lead to mechanical integrity issues. A systematic process of such periodic inspections is essential to provide early detection of beginning problems, and permit timely corrections. I can give several examples of potentially catastrophic events which were avoided by this inspection process in chemical process units under my supervision. Failure to include such inspections is a serious flaw in CPI safety programs. Many experienced operating managers conduct such inspections, but safety consultants rarely include them in their litany of safety program recommendations.

J. Dale West, PE
[email protected]

Author Bill Mostia, PE, replies:

I agree it would have been nice to provide the underlying causes in my table. The problem with writing an article for a magazine like Control is I have limited space and underlying causes can't always be boiled down to a few words. It also takes a lot of investigation and information not always in the public domain. As it was, I started out with an article that was about twice the size as the final version and I had to whittle it down to fit into Control. The magazine is nice enough to allow me to write longer than average articles as it is. Overall, the analysis of underlying causes would probably fit better into a book than an article.

The causes of the Bhopal incident are open to some debate depending on who you believe. As is not uncommon, company involved, Union Carbide India, had its own investigation, which was favorable to its legal case (internal sabotage), while other people make the case that Union Carbide India was lax in its safety duties, which is favorable to their case. It's not uncommon for vested interests to color the investigation of accidents on both sides of the argument, particularly when large amounts of money are involved. The answer almost always lies somewhere the middle.

In any case, the process safety technical systems should have made it difficult for the accident to occur and/or mitigate it to some extent. The likely initiating event was sabotage, but that's just the beginning of the story of how the accident progressed, and the question arises as to whether there were engineering safeguards and administrative procedures in place to help prevent or mitigate the accident, and what was the response of the people as the accident progressed (e.g. what systematic errors facilitated the accident).

In regards to inspection being important, I fully agree with you, but many times a lack of money or resources prevents or interferes with having an adequate inspection program. Inspection and testing are cornerstones of a good mechanical integrity program, which is key to having a reliable plant. Consequently, a reliable plant translates into a safe plant.

If you wish to discuss other aspects of process safety, I'd be happy to do so. Please note that I am not an OSHA PSM guy. My expertise lies more in safety instrumented systems (SISs), technical systems, loss prevention (LOPA) and safety reliability.