Friday, July 20, 2012

Cognitive Dissonance at Palisades

“Cognitive dissonance” is the tension that arises from holding two conflicting thoughts in one’s mind at the same time.  Here’s a candidate example, a single brief document that presents two different perspectives on safety culture issues at Palisades.

On June 26, 2012, the NRC requested information on Palisades’ safety culture issues, including the results of a 2012 safety culture assessment conducted by an outside firm, Conger & Elsea, Inc (CEI).  In reply, on July 9, 2012 Entergy submitted a cover letter and the executive summary of the CEI assessment.*  The cover letter says “Areas for Improvement (AFls) identified by CEI over1apped many of the issues already identified by station and corporate leadership in the Performance Recovery Plan. Because station and corporate management were implementing the Performance Recovery Plan in April 2012, many of the actions needed to address the nuclear safety culture assessment were already under way.”

Further, “Gaps identified between the station Performance Recovery Plan and the safety culture assessment are being addressed in a Safety Culture Action Plan. . . . [which is] a living document and a foundation for actively engaging station workers to identify, create and complete other actions deemed to be necessary to improve the nuclear safety culture at PNP.”

Seems like management has matters in hand.  But let’s look at some of the issues identified in the CEI assessment.

“. . . important decision making processes are governed by corporate procedures. . . .  However, several events have occurred in recent Palisades history in which deviation from those processes contributed to the occurrence or severity of an event.”

“. . . there is a lack of confidence and trust by the majority of employees (both staff and management) at the Plant in all levels of management to be open, to make the right decisions, and to really mean what they say. This is indicated by perceptions [of] the repeated emphasis of production over safety exhibited through decisions around resources.” [emphasis added]

“There is a lack in the belief that Palisades Management really wants problems or concerns reported or that the issues will be addressed. The way that CAP is currently being implemented is not perceived as a value added process for the Plant.”

The assessment also identifies issues related to Safety Conscious Work Environment and accountability throughout the organization.

So management is implying things are under control but the assessment identified serious issues.  As our Bob Cudlin has been explaining in his series of posts on decision making, pressures associated with goal conflict permeate an entire organization and the problems that arise cannot be fixed overnight.  In addition, there’s no reason for a plant to have an ineffective CAP but if the CAP isn’t working, that’s not going to be quickly fixed either.

*  Letter, A.J. Vitale to NRC, “Reply to Request for Information” (July 9,2012) ADAMS ML12193A111.

1 comment:

  1. The stance of the Vitale letter regarding the Conger & Elsea NSC Assessment results for Palisades also caught my attention.

    After any amount of time in pursuit of performance improvement, any statement by Senior Management that sounds like "we already know about that and have it covered in our Plan" will arouse suspicion.

    Any more, that suspicion can tend to go reflexively in the direction of "Oh, oh; they still don't get it." But I found myself wondering if that presumption might be a little more precarious than it was before the days of Safety Culture Assessments.

    I see two problems: 1) generic assessment mechanisms, and 2) mistimed assessments. My concerns emerge from anecdotal observations which I suspect could be incorporated into a systems theoretic representation, but let me just jump to some tentative hypotheses.

    1) Most of recent the large organization Nuclear Safety Culture Assessment architectures are traceable to an organizational model described in NUREG/CR-5538 "Influence of Organizational Factors on Performance Reliability" which was developed for NRC circa 1991.
    2) These assessments are intended to be acutely diagnostic - they have little longitudinal performance aspect; they encompass a wide range of techniques but the results are typically dominated by written individual survey responses that are arrayed on a fitness continuum (e.g. a seven point Likert scale).
    3) When applied to an organization that has been under-performing in the view of an Independent Oversight body for some time, the assessors are already looking for a breakdown in trusting relationships within the enterprise - not surprisingly they find it.
    4) Reports of pervasive weakness in relationships are typically rendered in language sanitized of local context (cf. the C&E Executive Summary) - this is intended to suggest that the report is "objective."
    5) These generic indictments of senior management can be taken as evidence of the Fundamental Attribution Error in the assessment process - the recipient has no ready internal means of validating what "meaning" to attach to the findings - but they are compelled to "accept the assessor's view" or refute it.
    6) If the organization, like Palisades (or BNI/WTP), is already under increased oversight scrutiny, the conduct of an Independent Assessment well into the process of a recovery plan is likely to create dissonance just by virtue of ill-timing. Of course some, if not all, of the issues surfaced in the assessment will seem familiar - particularly when presented in language that is very generic and has little provision for temporal nuance.
    7) The assessments provide little indication if the process of rebuilding relationships is begun or even an actual objective of the several stakeholder parties (workers can make or break a CAP all by themselves).

    Tentative Conclusion: If the circumstance at the Hanford WTP is indicative, multiple NSC assessments, performed by different groups, only a few months apart, and with no common theory or method, and after a systematic effort to improve performance has begun are a really bad idea. They do more harm than good.


Thanks for your comment. We read them all. The moderator will publish comments that are related to our content.