Model based systems engineering breaks down silos and allows real engineering teamwork.
Episode Summary:
The design and manufacturing of complex projects has always been a challenge in engineering. Several novel management techniques evolved over the last 60 years to cope with the problem of keeping projects on time and on budget, but early tools rarely tackled the problem at its root cause: the lack of truly collaborative engineering design. Documents are necessary in engineering, but they have had the effect of handcuffing engineering teams and de-incentivizing real creativity. Model based systems engineering promises to remove those handcuffs and profoundly improve product and project performance.
Access all episodes of End of the Line on Engineering TV along with all of our other series.
Transcript of this week’s show:
To see any images, graphs, charts, graphics, and/or videos to which the transcript may be referring, watch the above video.
One of the best parts about working here at engineering.com is the opportunity to meet so many smart, smart people in industry. Now software has always had smart people, but the traditional intersection between software and engineering, CAD/CAM, has morphed into something much bigger and much more important than most people realize. I was a panelist at a recent roundtable hosted by Dassault Systems, their 3DExperience Forum where I commented on the current state-of-the-art in software assisted engineering: Model Based Systems Engineering.
Now we’ve all been watching advancements in artificial intelligence, augmented and virtual reality, simulation and generative design, but these technologies, although important, dance around the fundamental core of engineering. Artificial intelligence does what human engineers do, presumably faster and with fewer errors. Augmented and virtual reality are the logical extensions of two and three-dimensional renderings. Simulation is the virtualization of hardware-based R&D, and generative design Is simply the automation of what every design engineer does every day in developing new parts and assemblies. They’re all essential, but they don’t fundamentally change the way engineering is done.
Model based systems engineering, however, is different. This is about removing the handcuffs that constrain every modern engineering project: documents. System complexity in engineering has always been a problem, and with the advent of commercial digital computers in the 1950s, major defense programs became too complicated to effectively manage. Determining the critical path on a large Gantt chart was the only way to handle most problems, and the need to police revision histories on thousands of individual part and assembly drawings, plus associated documents related to testing, regulatory compliance, quality and scheduling within major projects, basically guaranteed that they would come in late and over budget. Cost, delays and the root problem of getting dozens or hundreds of engineers’ designs to come together into subassemblies, assemblies and final products has been a perpetual issue, whether the project was an electric toothbrush or an Apollo moon rocket. Why?
In my opinion, there are two primary reasons: the first was and is simple: projects are costed relatively early in the design phase, and there is no way to effectively predict how much redesign may become necessary, or in the event of a poor design making it to production, eventual cost of rework. For large projects, especially in defense, the only way to get a big project across the finish line was either to price new projects at astronomical levels, or negotiate cost plus contracts, or both.
The second reason was and is simple: With each individual engineer responsible for one part or one subassembly of a large project, their motivation, and metric of success is determined by the functionality, cost and optimization of the single thing that they work on. Integration of that part into large complex assemblies like airplanes or automobiles, was typically handled by a different engineering team responsible for integration. The result was inevitable negotiation and horse-trading to get parts into subassemblies and then into assemblies and on into products. All of it costs time and money.
The brilliance of model-based systems engineering is that a single digital concept of a large and complex product is accessible to everyone, at every level of product development. An engineer may be responsible for a flange or bracket, for example, but he or she has the ability to see how it fits into the larger project, and also sees what the ripple effects are of engineering changes all the way through the process, from design revision to manufacturing. This is incredibly powerful and adds a real incentive for engineers to think holistically about every step of personal design process. Want to add a lightning hole to take mass out of that hypothetical bracket? What if that design change affects the flow of cooling air past a circuit board?
With model-based systems, difficult to predict outcomes can be simulated and revealed before production begins. And with the transparency of the process everyone can see the impact. In this simple example, a change to another component or assembly might allow that hole without compromising performance and reducing cost at the same time. But with engineers working in silos, with traditional document-based systems, there is no way to know how a small change can have big system benefits, so sub optimal designs are almost guaranteed. Model based systems engineering has other benefits and will be talking about the more here at engineering.com.
And by the way, visit the Dassault Systèmes 3DExperience Forum if you’re interested in hearing what I had to say in detail. In the meantime, think about model-based systems, and think hard about how much document-based systems put engineering professionals in a straitjacket.
