General lessons from incidents

At JACADS and TOCDF, operations personnel did not appear to generalize lessons learned beyond the immediate equipment and task in the original incidents. There is room for making much wider use of these valuable lessons, such as by mining the information in the PLL database to detect patterns that may underlie several incidents. The effort to prioritize the data is a good start toward increasing the information s usefulness. PMCD could also make better use of information available from industries such as the chemical and petroleum manufacturing sectors. Both have... 

The chemical industry generally handles business so well that it is difficult to find large numbers of recent incidents for examples. Many of the featured case histories in this book occurred over 20 years ago however, the lessons that can be learned will be appropriate into the twenty-first century. Tanks can fail from the effects of overpressure and underpressure in 2010 just as well as they failed in the 1980s. Incompatible chemicals are incompatible in any decade and humans can be forgetful at any time. Before we review a single case history, it is time to boast about the safety record of the chemical industry. 

Such are the results and lessons of 1997. As to general results and lessons learned, they are rather trivial. The main conclusion we arrived at is that it is necessary to carry out measures preventing accidents. Analyzing the incidents of 1997, we can see that the preventive measures prove to be more economic than the elimination of accident consequences. We must always remember the lessons of Chernobyl, and these lessons show that in Russia today the annual sum of social security payments to those who participated in the elimination of consequences from the Chernobyl accident and to the population affected by this accident is six trillion roubles (in old prices), i.e., 1 billion. So we spend annually 1 billion for social compensation. Thus, the prevention of accidents is a routine principle, which we must observe in Russia today. 

In each incident you should ask yourself (1) What happened (2) How did it happen (3) Why did it happen This series of questions is one form of root cause analysis (RCA), a standard procedure in the investigation of incidents in business and industry. The last question is the most important one and you will find you may have to ask and answer this question several times in sequence before you come close to the real or root causes of an incident. RCA generally recommends asking Why five times to get to a fundamental, rather than superficial, cause. Root causes are the basic causes of an incident that can be reasonably identified, that can be controlled, and for which recommendations or lessons learned can be derived. Many times root causes are not immediately obvious, but can be identified from careful inquiry. Once you know the root causes you should be able to develop recommendations or steps to prevent this from happening again. 

---