Disaster by Design/ Safety by Intent #62
Safety by Intent
The Nuclear Regulatory Commission (NRC) revised its regulations requiring nuclear plant workers to be fit for duty on March 31, 2008, to include measures intended to protect against mistakes made by workers impaired by fatigue. Specifically, Subpart I, “Managing Fatigue,” was added to 10 CFR Part 26, “Fitness for Duty Programs.”
The NRC issued its fitness for duty requirements in the late 1980s in response to Congressional concerns about reports of illegal drug use by nuclear plant workers. The original requirements instituted initial, random, and for-cause drug and alcohol testing of all workers granted access inside nuclear power plants.
On September 28, 1999, Barry Quigley, a worker at a nuclear power plant in Illinois, petitioned the NRC to revise its fitness for duty regulation to impose limits on the number of hours worked by nuclear plant staffers. Quigley proposed that the NRC revise its regulations to limit workers to 60 hours per week and 108 hours over al-two-week periods during normal operation with some relaxation of these limits when the plant is shut down.
The NRC convened a series of public meetings to discuss the working hour limits petition. I participated in many of the meetings on behalf of UCS. The matter interested a lot of groups and individuals with a range of viewpoints. Some, like the Professional Reactor Operator Society (PROS) and the International Brotherhood of Electrical Workers (IBEW), expressed concern that limits on working hours could reduce the amount of overtime pay earned by their members. They pointed out potential consequences from the working hour limits being sought—workers might obtain second jobs to replace the lost overtime wages.
The NRC invited specialists to the meetings to inform the discussions. For example, Dr. Gregory Belenky from the Division of Neuropsychiatry at the Walter Reed institute of Research presented insights from research on sleep and human performance.
Fig. 1 (Source: Nuclear Regulatory Commission)
The differing perspectives by various stakeholders shaped by the insights from the specialists led to changes in the working hour limits proposed in Quigley’s petition. For example, the petition would impose working hour limits on all persons subject to the drug and alcohol provisions in Part 26. Industry representatives pointed out that many workers performed tasks that were very unlikely to contribute to a nuclear plant accident even if a worker was impaired by fatigue during their performance. An engineer calculating the power supply loads to be met by an emergency diesel generator could make a mistake due to fatigue. But that engineer’s work must be independently checked by another engineer and approved by a third individual, providing opportunities for any mistakes to be corrected. The scope of the fatigue management rule was therefore narrowed to only those workers with “hands on” tasks whose mistakes might more immediately and directly have adverse consequences—operators, maintenance personnel, and (after 9/11) security force personnel.
The Outcome
The final rule limits individuals to working no more than 16 hours during any 24-hour period, 26 hours during any 48-hour period, and 72 hours during any 7-day period.
The final rule also contained measures intended to address the situation where individuals are within the working hour limits but feel fatigued. For example, a worker may not have gotten a full night’s rest due to caring for an ill child. Under the final rule, that individual has the right to self-declare feeling fatigued without fear of reprisals. This provision explicitly addressed a problem that NRC identified in 2002 with some workers being compelled to work when impaired by fatigue or being fired for refusing to work.
The final rule contains another provision that had not been in Quigley’s petition that was added as a result of the insights provided by the specialists—the need for break periods. The specialists agreed that it was vitally important for individuals to have periodic opportunities for rest and relaxation. The final rule therefore requires that workers receive a 10-hour break between work periods and a 34-hour break every nine days. Individuals working 8-hour shifts must be given at least one day off per week, persons working 10-hour shifts must be given at least two days off per week, and individuals working 12-hour shifts must be given at least 2.5 days off per week when averaged over a shift cycle of six weeks or less.
The final rule recognized that situations could arise that required individuals to exceed the working hour limits. The rule contains a provision describing when management can authorize individuals to exceed the limits. The final rule also requires owners to submit annual reports to the NRC on the number of times when individuals exceeded the working hour limits.
The final rule adopted by the NRC was more compromise than consensus. No stakeholder got everything he or she wanted included in the final rule. I am not a fan of the provision allowing individuals to work longer hours when a reactor is shut down, particularly when that reactor is at a plant with multiple reactors and at least one other reactor is operating. But the NRC’s rulemaking process considered my viewpoints equally with the viewpoints of Barry Quigley, PROS, IBEW, the nuclear industry, and all other stakeholders. The NRC’s regulatory analysis supporting the final rule explained why some aspects were included in the final rule while others were not.
Many NRC staffers were involved in the multi-year process leading to the final rule. Among them, Dr. David Desaulniers played a key role as the efforts’ technical lead. In addition to the considerable education and experience he brought to the table, Dr. Desaulniers served as a calming influence when stakeholders passionate about their viewpoints disagreed on the problem and its solution. He effectively enabled open discussions while discouraging non-productive debates.
The final rule does not provide absolute protection against mistakes made by fatigued nuclear plant workers. But it provides significantly better protection than had existed.
Disaster by Design
The problem that happened yesterday is much easier to solve than the problem that might occur tomorrow.
Worker fatigue did not cause the Three Mile Island accident and did not contribute to the Davis-Besse near-miss. Worker fatigue was therefore a potential problem. Nevertheless, the NRC revised its regulations to better manage the risk from this potential problem.
No one can count the umber of accidents avoided by the NRC’s working hour limits rulemaking. But the NRC’s rulemaking makes it less likely for worker fatigue to account for one accident. That seems like a good deal.
—–
UCS’s Disaster by Design/ Safety by Intent series of blog posts is intended to help readers understand how a seemingly unrelated assortment of minor problems can coalesce to cause disaster and how effective defense-in-depth can lessen both the number of pre-existing problems and the chances they team up.