Nuclear Power Safety: Numbers and Context

December 9, 2014 | 6:00 am
Dave Lochbaum
Former Contributor

Nuclear Energy Activist Toolkit #44


So what? Even if it is the best number as Dr. Sheldon Cooper claims on “The Big Bang Theory,” it’s just a number unless it is placed in context.

If it is someone’s age, it tells us something. But is it telling us the person’s age in years, or days?

If it is someone’s bank balance, it tells us something. But is it telling us the person’s balance in millions or in cents?

If it is someone’s golf score, it tells us something. But is it telling us the person’s tally for the first hole, the front nine, or the entire 18-hole course?

Finding the proper context for nuclear numbers is challenging, but the exercise is needed to ferret out as much context about the numbers as can be ferreted.

Consider reading in a Nuclear Regulatory Commission (NRC) report that Nuclear Plant A had 352 safety problems identified while Nuclear Plant B had 18 safety problems identified (Fig. 1). Absent any other context, these numbers might suggest that Plant B is the safer plant because it had fewer problems.

Fig. 1.

Fig. 1 (click to enlarge)

Consider finding on a later page in the NRC report that Plant A’s problems had been identified over a 39-year period while Plant B’s problems had been found in just the past year. The context added by this insight might now suggest that Plant A is the safer plant, experiencing half as many safety problems on an annual basis as Plant B (Fig. 2).

Fig. 2.

Fig. 2 (click to enlarge)

Consider learning from another NRC report that Plant A’s problems were identified during the ten-year inservice inspections while workers identified problems at Plant B during monthly inspections. Now, Plant B appears to be the safer plant, experiencing about one-tenth of the number of safety problems per inspection of Plant A (Fig. 3).

Fig.3 (click to enlarge)

Fig.3 (click to enlarge)

But the ten-year inservice inspections examine many more components than the monthly tests, offering many more opportunities to find problems. Suppose that each ten-year inspection covered one-quarter of the components without overlap such that the four inspections examined every component. And suppose that each monthly inspection examines 5% of the components. Even with no overlap (i.e., checking the same component(s) in more than one monthly inspection), the past dozen monthly inspections would have only examined 60% of the components (Fig. 4).

Fig. 4. (click to enlarge)

Fig. 4 (click to enlarge)

Wait for it, wait for it.

Assuming that all identified safety problems were adequately fixed and that the fixes including appropriate measures to preclude the components failing again, Plant A turns out to be safer because it is not likely to have any problems remaining to be found and fixed. Plant B, on the other hand, may have another dozen safety problems to be identified once the remaining 40% of the components get inspected (Fig. 5).

Fig. 5 (Click to enlarge)

Fig. 5 (Click to enlarge)

Bottom Line

Finding the proper context for nuclear numbers is challenging.

If one only had the numbers in the first column, it would be hard to figure out which plant was safer. Plant A clearly had more reported problems, but it would also have more safety fixes.

The quest for context seeks to gain insights as accurate as possible for the numbers in the last column. For nuclear plant safety is not dictated by the number of identified problems—it’s driven by the number of unidentified, and thus unresolved, problems.

The late Dr. Pietro Pasqua, head of the nuclear engineering department at The University of Tennessee when I pursued my degree, taught me that having all the right answers remains insufficient until all the right questions are asked. Dr. Pasqua’s lesson also applies to having numbers without proper context.

It may be impossible to ask, yet alone answer, all the right questions about complex issues or to obtain the proper context for every nuclear number. But the more successful the effort to reach these goals, the better the outcomes from decisions made on the partial information.


The UCS Nuclear Energy Activist Toolkit (NEAT) is a series of post intended to help citizens understand nuclear technology and the Nuclear Regulatory Commission’s processes for overseeing nuclear plant safety.