Insights From a NAFEMS Technical Fellow

Stating that numerical analysis has evolved over the years is kicking in an open door, as the Dutch say. Nevertheless, a constant factor over the decades has been concern about the quality of the product delivered by the analysis. I am not talking about the tools here, but the hands and brains that handle the tools. This concern dates back to the early days of desktop computers when Finite Element Analysis became available within the office, and the dependency on large mainframes weakened. Up until then, the use of the method was restricted to a limited number of people, considered some kind of wizards that produced a pack of punch cards that was fed into a machine large enough to fill a standard living room. The first cards in this pack contained mysterious instructions, dictated by the Job Control Language. Bearing all that in mind, imagine the kind of mayhem that might occur when power is placed in the hands of a sorcerer’s apprentice, and you get the picture.

In the second half of the eighties, this concern was addressed by a working group initiated by CIAD. This Dutch association didn’t survive, but the heritage has not been forgotten, at least not by me. A starting point was the development of benchmarks to familiarise the novices with the tools. The working group’s attention was soon drawn to NAFEMS, then a British national agency working on finite element methods and standards. As a result of all the efforts of the CIAD working group, Nevesbu (originally an abbreviation of Nederlandse Verenigde Scheepsbouwbureas) became a NAFEMS member in the early nineties. Incidentally, like NAFEMS, Nevesbu shook off the original basis of its name and lives on to celebrate its ninetieth anniversary next year.