Saturday, May 14, 2005

Safety and fatalities

From the May 9 LANL NewsBulletin:

In the discussion of Lab safety, the topic of fatal accidents at work has not been discussed. Given that the Lab employment is on the order of 10,000 workers and that many years of data are available, it is a statistically significant sample. This also is timely as we are past the 10th anniversary of the tragic training exercise shooting. Since that tragedy we have recorded approximately 100,000 man-years worked without a fatality. The Occupational Safety and Health Administration publishes its statistics in the same units, fatalities per hundred thousand man-years worked. Collecting the data together with the Lab data it is clear that our fatality rate is below that found in the educational and financial industries. The work at the Lab consists of a great deal of paperwork (too much for our own good), as with the lowest risk industries, but also consists of significant manufacturing and construction work. Compared to those two industry segments the Lab has been very successful at preventing fatal accidents.

Given also the nuclear weapon materials, the high explosives, the ubiquitous megawatt size electrical power sources, the occasional surprise receipt of live anthrax (through errors of external parties), and the armed guard force, this is a extraordinary record of worker safety. The record is so outstanding that it must be credited to a combination of factors: the safety intiatives that followed a series of severe accidents a decade ago; widespread compliance with those initiatives by the work force; and an additional quanta of good fortune.

One can only be bewildered by the widespread criticism of University of California management and the Lab work force found in the media, and be deeply concerned about the impact of the inevitable.

Table. Fatalities per hundred thousand workers by industry segment* or employer**.

Los Alamos and Cont. 0
Agriculture 31
Construction 12
Education 0.8
Financial 1.4
Manufacturing 2.5
Mining 27
Transportation 17

*OSHA Census of Fatal Occupational Injuries (2003)
**Los Alamos data for 10 years (~100,000 man-years)

--David Reagor



Comments:
The outside criticism is just parroting back what Nanos and upper management had told DOE, the press and Congress. If they told them the truth then they would not be given the power to be a "change agent", power, thats why the lies could not be retracted.
 
Actually, BLS publishes stats in fatalities per 100,000 worker units for a given year. This is an important distinction. Your table is therefore misleading as there were not 100,000 workers at LANL this year. Another way to put it, in case you think I am being pendantic, is that this is cherry-picking. If this year, god forbid, someone is killed on the job at LANL, the rate will shoot up to about 10/100,000. Not something to brag about anymore.

I generally agree with your point, that LANL is a very safe place considering all the hazards, but be careful about the statistics of small numbers (which is why people generally stay away from such subjects). You can quickly induce the trap of zero tolerance safety policies, which leads to absurd behavior like laboratory shutdowns. Do you want the lab to be closed the next time a careless driver runs someone over, or a deranged idiot shoots someone? Do you want managers who honestly believe they can create an environment where the above incidents will NEVER happen?
 
How can a manager honestly believe that he or she could create an environment where some deranged person would never go postal or an environment where human beings never cause a human error. Managers should work to put in place engineered solutions to lessen the consequences or probability of human errors; ensure that the staff have the appropriate as opposed to check-the-box training; and empower staff and team leaders to be elements of the solution and not persons to be suspected and watched.

Otherwise, I agree that statistics that are your best friend today might be your worst enemy tomorrow.
 
Go visit Efren Martinez, who lies in a nursing home (Sombrillo) after the electrical accident in Jan, 1996. Then read the accident report, which clearly indicts two LANL managers for having violated LANL safety policies and procedures.
The two managers were not punished; they were given large raises and promotions. The managers were exempt from lawsuit because the work for UC, which has protection from suits because it is part of a state government. LANL settled the lawsuit by Efren's family for $13 million.
This happened under Hecker and Browne. It is important to watch not only the "numbers" but also the follow up to accidents. The Martinez case, and others, sent a clear, and wrong, message on safety issues. It is time for a different message.
 
Hello, My safety letter was originally posted on the reader's forum at LANL. Someone posted it here and I have received many thoughtful comments both electronically and in private. Here are my responses to the comments posted on this blog.

Response to anonymous#1-
Excellent point. Pretend everything is fouled up to give the impression of being a "swamp drainer".

Response to bystander-
The lab safety statistics are good in all serious categories. Look at T/T (OSHA talk for terminations and transfers, injured workers that cannot return to work). DOE had 14 last year and 0 from LANL. The DART and TRC scores reflect ergonomics and bad footing. If they collected hospitalizations data LANL would likely do well. You can get that impression by looking at all the serious accidents on the DOE investigations website. The actual LANL safety performance is outstanding.

Response to anonymous #2-
This comment reflects the point of the final sentence of my letter. What kind of response would a actual serious accident beget?

Response to anonymous #3-
There were some very tragic accidents in the mid-1990's. Often the accidents were accompanied by significant mistakes by others. The safety systems were greatly improved after that time. The record between that era and the present is, on the average, outstanding.
 
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?