Last March at the National HPCC Conference, in Newport, RI, Caterpillar's Keven Hofstetter described virtual product development at the company — an impressive effort that's been going on for more than 20 years.
However, despite its success, the program is not without its problems. In the section of his talk titled "A Few of the Hurdles," Hofstetter cited "cultural issues" as one of the major stumbling blocks to user acceptance of modeling and simulation (M&S).
For some Caterpillar engineers and designers (and it sounded like there were quite a few), the old way — physical prototyping — was essential to getting product out the door. You can test a physical prototype, they say, and get real data; you can feel the iron; you can trust the results. For them, simulation on a computer screen was just not real, and was, at best, suspect.
A few weeks ago I was talking with the manager of a fitness club who, in a previous incarnation, was the supply chain manager of an electronics firm. When he asked me what I did, we got into a conversation about digital manufacturing. When I mentioned reducing dependence on physical prototyping, he threw up his hands. "No, no," he said, "You have to make and test real prototypes. It's the only way you can actually tell if your product is going to work — you have to hold it in your hands."
Proceed with Caution
Despite the fact that the adoption of digital manufacturing modeling and simulation can reduce costs, improve products, shorten time-to-market and, in general, boost a company's competitive capabilities, rapid and whole-hearted adoption by the several hundred thousand small- to medium-sized manufacturers (SMMs) in the US is not guaranteed.
In addition to a distrust of digital in lieu of physical prototyping, the Alliance for High Performance Digital Manufacturing (AHPDM) has identified a number of other hurdles that must be overcome before M&S makes major inroads into the SMM community.
- Tools — Today's M&S tools are hard to use and highly specialized. Analysis software must be adapted to support a particular manufacturing process. Even though the cost of HPC clusters has dropped dramatically, the expertise needed to install, configure, run and maintain even a small cluster requires IT skills that many SMMs just don't have.
- Talent — According to AHPDM, M&S remains a niche activity that "attracts relatively few computer scientists and engineers." Computer scientists and engineers who are skilled in M&S technology tend to gravitate to the large government labs and major corporations.
- Integration — In many cases, highly-knowledgeable computer scientists are required to employ M&S tools and then interpret the results to other teams of manufacturing specialists who in turn integrate the new data into equally complex, specialized processes and workflows.
- Deployment Risk — For SMMs with tight budgets and limited people resources, adding M&S to existing manufacturing processes is a risky business. They need all the help they can get to transition to processes that allow them to design and test in virtual space and minimize the need for fabricating physical prototypes.
Help for the "Missing Middle"
Happily, help is on the way. Organizations like the National Center for Manufacturing Science, AHPDM, the Council on Competitiveness, as well as many federal and state agencies, are addressing the needs of the "missing middle" and crafting a variety of unique solutions.
Part of the Digital Manufacturing Report's editorial mission is covering these multifaceted efforts, which have the potential to revolutionize manufacturing in the United States. We will also report on SMMs that have taken the plunge into the world of digital manufacturing, modeling and simulation, and have not only survived to tell the tale, but are reaping the rewards as well.