Data-driven decisions

Although we have not really had much sunny weather this year, it has been a particularly good summer for sport. My highlight has been the Olympics, watching individuals and teams chasing their dreams of a medal, ideally gold. The message spelt out by the winners is that dreams do come true, so chase your dream. Meanwhile, those who finish outside the medals have the consolation of calling themselves Olympians. They also have to decide whether they have the motivation and wherewithal to continue chasing their dream for another four years.

For as long as I have been aware of the Olympics, there has been talk of the Olympic dream and its realisation, just as there is the dream of any sporting achievement, career choice or lifestyle. As is the tendency these days, there is a need to repackage the old, rename it and make out that it is something new. This is ‘manifestation’, a noun that was new to me, although I am familiar with manifest as a verb. Two examples were interviews with sports people in the run up to the football Euros and the European Athletics Championships. Manifestation is a sign of something existing or happening. To manifest can mean to create something or turn something from an idea into reality. In psychology, manifestation is using thoughts, feelings or beliefs to create a physical reality.

Recently, the use of another unfamiliar term had me typing into the search engine: data-driven decision-making (DDDM). As you might have guessed, DDDM is the use of data as well as facts and metrics to inform strategic business decisions, as opposed to using guesswork or solely relying on intuition and observation. Although I am aware that these latter three options have been used in decision-making, I struggle to think of many decisions where data has not played some part in the final output. Of course, the rise of DDDM is the result of advances in technology and the ability to search through vast amounts of data to identify patterns, which can better inform decisions than human analysis alone.

Non-destructive testing (NDT) has always used DDDM to identify and sentence defects within components. NDT is not an absolute measurement but relies on the analysis of data collected by various NDT methods and techniques to decide whether there is a flaw and, if so, whether it is acceptable or rejectable. What is new is the amount of data that advanced techniques can generate.

Manifestation and DDDM are simplistic terms that belie the processes that have to be undertaken in order to produce the desired outcome. Brent Dykes^[1] highlights eight pitfalls of the DDDM process. Written with business in mind, they can also apply to the NDT DDDM process. The first is bad data. There are many steps that have to be followed to ensure that good data is collected. The correct NDT technique has to be selected. The parameters need to be specified and controlled during application; any deviation can have a big impact on the quality of the data. The procedure needs to be followed by operators who understand the importance of applying it precisely.

The second relevant pitfall is weak analysis. The collection of good data is just the first step in the inspection process. This data requires investigation to identify the presence of abnormalities, which can be used to infer the existence of a defect. The analysis has to be based on knowledge of the data patterns caused by defects that could potentially be present. However, as we have all experienced, such patterns are not always clear cut and a robust analysis can only be achieved through problem-solving and the collection of more data.

The final pitfall I want to highlight is the lack of learning taken following DDDM. There is a tendency to make the decision, heave a sigh of relief and then move on to the next job. This misses the opportunity to extract extra benefit from the inspection and the decision-making. By comparison with previous inspection data, trends can be identified that will provide additional information on the asset. Other operators can observe the data and the decision process and gain valuable experience and knowledge. The data collection and analysis procedure can be reviewed and any potential improvements identified and implemented in future inspections.

Manifestation and DDDM are easy labels that gloss over challenging processes that need to be applied to deliver results. Until such time as NDT becomes an Olympic sport, the NDT profession needs to concentrate on harnessing DDDM for maximum benefit.

Reference

B Dykes, ‘8 pitfalls in the data-driven decision-making (DDDM) process’, 2022. Available at: www.forbes.com/sites/brentdykes/2022/08/31/8-pitfalls-in-the-data-driven-decision-making-dddm-process

Please note that the views expressed in this column are the author’s own personal ramblings for the purpose of encouraging discussion within NDT News. They do not represent the views of Jacobs or BINDT.

Letters can be mailed to The Editor, NDT News, Midsummer House, Riverside Way, Bedford Road, Northampton NN1 5NX, UK. Email: ndtnews@bindt.org or email Bernard McGrath direct at bernard.mcgrath1@jacobs.com

Comments by members

This forum post has no comments, be the first to leave a comment.

<< Back