The disaster in Japan will be studied for years. There will be more than enough blame passed and fingers pointed at people, companies, agencies, and the government responsible for the meltdown at Japan’s Fukushima Daiichi of the nuclear plant. But could it have been prevented?
An article last week on an MSNBC blog shed some light on the answer.
Rarely does one mistake or event cause a catastrophe. Disasters like what happened in Japan, in New Orleans, or at the World Trade Center on that fateful day 10 years ago are rarely the result of a single point of failure. (Could anyone have anticipated the two towers crumbling following the crashing of planes into the buildings?) They are the result of multiple mistakes and bad luck.
In Japan’s case, first came one of the most powerful earthquakes ever recorded. This was followed by a tsunami. The tsunami knocked out the backup generators whose batteries unfortunately were not designed to provide more than a few hours of coverage. The earthquake and tsunami wiped out the electrical grid that was supposed to back up the back up plan. The disaster plan never accounted for a disaster of this magnitude and duration.
Hindsight is always 20-20. Mistakes will be identified and corrected. But could any changes really have prevented this disaster? Could lessons learned prevent a similar catastrophe in the future?
Unfortunately it’s not likely. We live in a world of ever-increasing interconnectedness. The number of permutations in scenario thinking is mind-boggling. Even if plans could be designed to prevent or interrupt a cascade failure such as happened in Japan, implementation of preventive plans isn’t likely to happen. Human nature tends to want to fix things that are broken. The rush one gets from troubleshooting is far more rewarding than the satisfaction of preventing an event that many believe will never occur.
There's also a simple reason backups work well in theory but often fail when they encounter real-life trouble. According to Peter Neumann, principal scientist at the Computer Science Lab at SRI International, "It's impossible to simulate all the real things that can go wrong. You just can't do it…You can't anticipate everything."
Even if Neumann is wrong and you could account for all the possibilities, testing the scenarios is difficult. Simulating the pressure and humans emotions of that catastrophic moment is just not practical.
In addition, the ability to flawlessly execute such a multitude of diverse number of solutions is unrealistic. To train and reinforce the necessary skills would require an enormous amount of time and resources. And what happens when a key member of team is sick or on vacation; or the most experienced employee retires and is replaced by a talented but not so experienced worker. You just can’t account for all the variables.
Companies don’t invest enough time, money, or effort in the most basic preventive maintenance. They cover the most likely scenarios and place a bet that the worst won’t happen – at least in their lifetime.
At the same time, building redundancy into systems makes them far more complex, adding to more maintenance headaches and requiring even more skilled and experienced people to decode and operate. To adequately prepare for the once in a lifetime or even the once every 50 years scenario isn’t going to happen.
Finally safety costs money. One ugly reality of safe systems design is cost. No organization regardless of the risks is immune to budget constraint. Despite comprehensive cost-benefit analysis at the end of the day, the experts will be forced to make educated guesses…and they will be continually second-guessed. When additional funds are invested for prevention but no accidents occur, blame will rain down upon the company for over-spending. When a failure occurs, the innocent will be guilty of cutting corners.
The post-mortem on the disaster in Japan will last for years. We will likely learn about mistakes made– intentional and incidental. Regulations and procedures will be passed to prevent another tragedy. But risk management is a balancing act and many of the new laws and policies will be nothing more than window dressing. Risk management is as much art as it is science. And the science used is often inexact.
The inevitable questions become: What lessons will we learn from this disaster that will really make a difference?
Comments