This website does not use cookies. Read Privacy Policy here.
Beville Engineering Logo

Lessons Learned About Plant Operations In the 20 Years Since the Three Mile Island Accident


Remember back to March 28, 1979. Jimmy Carter was president, Disco was hot, and so was the reactor core in Unit 2 of the Three Mile Island nuclear plant - way, way too hot. It is hard to believe that twenty years have passed since the Three Mile Island Nuclear accident. The accident is probably the most studied industrial accident in history, and much of the blame for the accident was placed on poor human factors and inadequate man-machine integration. Two decades has given sufficient time for pause and reexamination of the accident and its aftermath. There are a number of important lessons to be learned in plant operations for those willing to take notice.

In observation of the twenty years since the accident, the Human Factors and Ergonomics Society interviewed Edward Fredrick, one of the two control room operators who was on duty as the events of the accident happened. The interview is presented in the April 1999 issue of Ergonomics In Design. The article is interesting, as it gives an operator’s perspective on the events contributing to the accident and the improvements that have been implemented in response. Mr. Frederick was an ex-Navy nuclear technician at the time of the accident and has since earned a Masters degree in human factors engineering and is responsible for root cause analysis, training and event prevention at the Three Mile Island plant.

In recalling the accident, Mr. Frederick recounts how the “plant was behaving oddly quickly” and within the first fifteen minutes realized that something was seriously wrong. At the time the plant procedures were event based, meaning that the event had to first be identified before a procedure could be selected and implemented. Unfortunately, the operators were trying to do pattern matching with instruments that were providing anomalous readings and it turned out to be a futile effort to find an appropriate procedure. After about one or two hours, the operators exhausted their options and switched to using engineering resources and other operators to diagnose the problems. In reflecting on diagnosing the event, Mr. Frederick said that switching to “symptom-based procedures was probably the greatest improvement as a result of the accident.”

Another point discussed during the interview was a Nuclear Regulatory Commission study following the accident that attributed much of the cause of the accident to anomalous instrumentation readings, the design of the control room, and operating procedures. Mr. Frederick disagreed with the conclusions, believing that the study missed the issue of training and experience and its contribution to the accident. As stated in the article, “If we had a different set of assumptions and a broader base of training, then how we interpreted the panels would have been different.”

After the accident, regulations were put in place that limited the types of control systems that could be used. All control systems had to be the analog panel mounted instruments; distributed control systems were not allowed. This limited the degree to which human factors could be applied in the control room. As Mr. Frederick pointed out, “Although we did some good things with meter markings and color coding and some additional mimicking and things like that, you are limited as to what you can do when backfitting a panel.” For the human factors improvements that were made in the control room, the observation was made that training on the improvements was lagging.

From Mr. Frederick's perspective, symptom-based procedures and increased training have been the biggest improvements since the accidents. Improvements to the control room helped, but were limited by federal regulations and thus have not had as much of an impact.

Copyright © 2000 Beville Engineering, Inc. , All Rights Reserved

For further information, visit the links below.

| Alarm & Display System Analysis Overview | Alarm Management | Display Design | Alarm & Display Design Seminar | Beville Shape Calculator -Excel download | Alarm Distribution Estimator -Excel download |


Article Published By
Consortium Reports New Findings on Alarm Rates Automation World
How Many Alarms Can An Operator Handle Chemical Processing
Impact of Alarm Rates and Interface Design on Operator Performance Automation World
Operator Interfaces: Moving from Comfortable to Most Effective Automation World
Operator Performance as a Function of Alarm Rate and Interface Design


This year's Fall meeting for the Center for Operator Performance will be October 7-10 in Houston, TX. For more information, please contact Lisa Via. Guests are always welcome!


Our summer newsletter is now available. Click here!

Take our short survey on operator span of control. Click here (new window)

David Strobhar's book, "Human Factors in Process Plant Operation," is now available in both hardcover and Kindle e-book.

Copyright © 1996-2019 Beville Engineering, Inc. All rights reserved. (937)434-1093.