NDT The Wisdom of Crowds

We are now well in to autumn and can no longer look forward to the predicted heatwave, which was always going to come next week but didn't materialise for any sustained period of time during the summer. On balance, the summer has been a bit of a disappointment, but there have been some glorious sunny days. One such day occurred on a Saturday and, not having anything urgent to do, I was able to saunter down to the local shops to pick up the morning paper. It was not exactly the break of dawn, but it was still that quiet period of the morning when the hazy sun's heat was just building and people were starting to stir and go about their business. I walked back home with a contentment inspired by the sunny day ahead, even though an urban pavement bordered by a row of parked cars is not the most glamorous environment for an ambulation. I then saw something that raised my spirits to an even higher level.

On the dashboard of one of the parked cars was a dancing duck! Not a real duck, obviously, because why would a real duck be dancing if the radio was off? It was a yellow plastic duck. The duck had a head, body and feet, with articulations between each so they could move independently. Even though the car was stationary at the side of the road and there was no one in it to cause movement, the duck was dancing. Observing a dancing plastic duck has the same effect as wearing red trousers: it makes you laugh. As the saying goes: you don't laugh because you are happy, you are happy because you laugh. So I laughed and, ignoring the disconcerted looks from passing motorists, I was happy.

In my happy reverie, the dancing duck led to two threads of thought. As I am not one to use up more than one thread per column, I am afraid you will have to wait until next month for the second dancing duck instalment. The reason why the duck was dancing was not immediately obvious. The car was stationary. There was no one in it. The windows were all dosed. The most obvious answer was that, despite the traffic being light, the air disturbance from passing cars was sufficient to shake the parked car and move the duck. I can't pass on the opportunity to say that the duck must have been lightly damped! However, there are other possible explanations: the disturbance from the passing cars could have been transmitted through the wheels and the ground; the sun may have set up thermal convection in the car, which impacted on the duck; or maybe it was a combination of two or more mechanisms.

Now, just suppose for a moment that, unlike all of the other occasions when I have asked for your input and received a deafening silence, you were to send me your suggestions of the likely explanation for the dancing duck. Due to the wisdom of crowds, the cause would likely be identified. Non-destructive testing has a long history of relying on the wisdom of crowds. All the British, European and international standards that the profession applies to inspections were drafted by people with experience in the field relevant to the particular subject matter. There are three main conditions necessary for a group of people to make wise, or intelligent, decisions. These are decentralisation, diversity and independence. The standards committees are generally made up of representatives from a variety of industry sectors, which will ensure a degree of all three attributes and will only increase when moving from national to international committees.

More recently, the wisdom of crowds has been exploited in expert elicitation panels. These are not dissimilar to standards committees and are convened when there is uncertainty about particular information that is needed to support decisions. They are applied in risk-based inspection (RBI) processes to identify the probability of defects occurring and also in the production of defect specifications, again to identify likely defects that might occur and their morphologies. The decentralisation, diversity and independence of these panels presents a challenge that needs to be carefully managed.

In the case of NOT, the committees of experts are focused on the detection of particular defects: standards specify inspection parameters in order to detect predicted, known defect types, although these defects are not necessarily defined in any detail and the outputs from RBI and defect specification panels provide the detailed inputs for the design of inspections.

The Achilles’ heel of any committee of experts is that infamous phrase: 'unknown unknowns: So, what happens when a defect, which has not been identified as probable, occurs? Fortunately, the answer lies in NOT! Some NOT should be applied to detect the unexpected. This is a difficult concept to grasp: why look for something if we do not expect it to exist? The answer lies in defining the shape and size of a defect that would threaten the integrity of the component, without worrying about how it might be generated, and then applying NDT to confirm that no such defects are present. The result is increased confidence in the safety of critical components.

Luckily, I am not so concerned with whether the wisdom of crowds would establish the cause of the duck's dancing, I am just happy that it danced.

Please note that the views expressed in this column are the author's own personal ramblings for the purpose of encouraging discussion within the NDT Newspaper. They do not represent the views of Amec Foster Wheeler or the HSE who funded the PANI projects. Letters can be mailed to The Editor, NOT News, Midsummer House, Riverside Way, Bedford Road, Northampton NNJ 5NX. Fax: 07604438300; Email: ndtnews@bindt.org or email Bernard McGrath direct at bernard. mcgrath@amecfw.com

Berard McGrath