Take a 3D look at your CAD/CAM strategy

Despite the advances of mid-range CAD/CAM software, with the achievements of products like SolidWorks, Solid Edge and Autodesk Inventor, little at the top has changed. The market still breaks down into big CAD and small CAD vendors, with the former going ‘enterprise wide’ and seeking to address business processes rather than mere geometry, while the small CAD vendors ever so slowly achieve 2D to 3D migration and supply pseudo enterprise capability with partners. Beyond that, collaborative engineering has taken over from concurrent engineering as the marketing buzzword du jour, with again little change, the only difference being the web! We’re simplifying, of course, and a little unfairly: specialists like Geometric Software Solutions, bom.com, CollabWare, Centric Software and RealityWave, will feel rightly affronted. Fact is, there has been progress, but only in the niches. Rounding out the observations, perhaps the most glaring is that there are still far too many companies using 2D systems when 3D solid modelling is patently the only sane strategy today. A recent Autodesk Inventor press release claimed 50% cost and time-to-market improvements using 3D as if this was something new: many have been achieving this and more since the late ‘80s with 3D. But Autodesk focuses rightly on what can be achieved precisely because so very many more have yet to be converted. And then there’s the old environment argument. Windows seems to be taking over the desktop while Unix is still firmly positioned for enterprise servers. Against this chequered backdrop how do you plan your IT strategy? If you’re still reading, you’re well on the way to achieving your first objective, which is to have a plan! Far too many IT screw-ups are the result of various departments ‘doing their own thing’ with little or no co-ordination by senior management – and engineering design is no exception. That’s not to say there is a universal CAD/CAM strategy: each company’s IT strategy should be as different as their respective businesses. So there is only one rule here: keep it simple. It sounds obvious, but go back to first principles and decide what you’re trying to achieve: in this preliminary analysis you should be speaking the language of business, not computers. Stable computing is key Nigel Hobden, CEO of HP system integrator Random Computing, says that with this done, a ‘stable computing platform’ is a key objective. “The stable computing platform is an approach to the design, specification and implementation of an IT infrastructure whose goal is to maximise the business benefits to the enterprise,” he says. “It approaches the specification of an IT system in a top down fashion,” he continues. “All too often the approach is to start with the hardware and work the specification up from there. The stable computing platform reverses this, in that it starts with the users and examines how their productivity can be maximised for the enterprise, and then from this baseline an optimum IT solution is developed.” Hobden identifies the three related issues of price, cost and value as the key considerations. “The stable computing platform is designed to maximise the value element while delivering at a realistic cost. “It is important to understand that [this] is built upon a set of technologies that will differ from solution to solution. It is not an ‘out-of-the-box’ approach but a considered solution developed from a proper understanding of the business issues.” However, to arrive at a value-based solution, he emphasises the need to understand the business characteristics of the IT components – hardware, infrastructure, operating systems, software and data. And since no one user company can be expert in all these, “It is essential to have an expert partner.” John Farrant, principal consultant with IBM Global Services, goes further. “With the rapid advances in IT, an ‘electronic’ engineering environment requires change almost as frequently as the products being designed; few product makers can keep up. Since managing a sophisticated IT infrastructure is seldom a core competency, manufacturers should consider entrusting it to an experienced provider, just as they rely on suppliers for product components.” The bottom line? Ralph Seeley, senior consultant at analyst Cambashi, agrees: “Many small companies cannot afford to employ IT professionals at all. Those that can, cannot afford to be anything but totally objective about what they spend their money on. If 3D modelling increases the bottom line, or improves the security of the business, or facilitates the winning of a big order it’s in. Otherwise it’s not.” And he adds: “It’s not just the cost of the system: it’s finding designers who have the aptitude to use it, training them and retaining them and then repeating the process every time a customer changes his/her mind about standards.” The bottom line for IT strategy? “If an SME supplier has to respond to the whim of bigger customers (who sometimes have more money than sense), he is advised to be light on his feet,” says Seeley. “Tails still do not wag dogs. That means for many, going for ‘lowest common denominator’ software. That’s why there’s so much AutoCAD LT. (And MS Windows for that matter).” But caveat emptor. On the subject of Lite and Microsoft, since the mid ‘90s the mere fact of using Microsoft Office and back office applications has been seen as good reason for selecting Microsoft-based CAD systems. Not any more. Sun Microsystems is attempting to make its StarOffice ubiquitous by giving it away (http://livingwithoutmicrosoft.org). StarOffice runs under Solaris, Windows and Linux and attempts to do most of what MS Office does: most importantly, it reads and writes MS Office files. Success is too early to determine, but it could reduce our dependency. It is also instructive to note that Microsoft uses Unix at the core of hotmail, msn and its own telephone system. The rights and wrongs of Microsoft operating systems are beyond the scope of this feature, but suffice to say that Microsoft’s tendency to update every 18 months is hardly conducive to a ‘stable computing platform’. No sooner have we sorted out Windows NT and opted for 2000 when XP comes along. And then there’s the three letter acronym (TLA) alphabet soup. Analyst CIMdata, at its European conference in Barcelona last year, focused on collaborative product definition management (cPDm). Definition: “a business approach to leveraging a wide range of technologies across the extended enterprise in programmes such as concurrent engineering, digital manufacturing, globalisation, supply chain management, integrated product development, e-commerce, and others.” So what happened to PDM (product data management)? Alstom Power, a Unigraphics user and a contributor to the conference discussed the company’s programme of achieving customer satisfaction through TLCPIM (total life cycle product information management). I’m not seeking to demean its excellent work, but this kind of activity does not help the many thousands still trying to escape the dark ages of the 20th century. The big CAD companies’ preoccupation with PLM (product lifecycle management) is only really relevant to about 1,000 companies world-wide and the continual redefinition of their technology simply leaves the rest of us even more confused. What we need is a return to basics and a strategy informed by them – and something to give us a kick start. Says Seeley: “Leading edge CAD software has reached the point of diminishing returns, so development in that direction is limited. Sooner or later, attention must turn to developing the market by popularising a product. We’re all still waiting for the ‘Windows 3.1’ of CAD. Or better still the GNU/3DCAD.” (see http://www.gnu.org) Brilliant innovation And that’s what’s just happened. All eyes are now turning to Autodesk which, with the advent last month of its combination Autodesk Inventor Series has stolen the limelight and opened the doors to total 3D CAD popularisation. Autodesk’s strategy is inspired: all its Autodesk Mechanical products, including 3D Inventor, are now provided on a single CD for a single price to a single user. The firm has protected its user base at no extra cost – and provided the potential for existing 2D users to migrate to 3D at a pace that suits them. Most importantly, Inventor Series offers wannabe 3D users a safety net. Other 3D CAD modelling vendors require users to take the 3D ‘leap of faith’ – abandon their existing 2D system and move to 3D with no escape route if it doesn’t work out. Yes, they can go back to their old 2D system, but that’s exactly what it is, old. With Inventor Series you get the latest version of everything. It’s win-win. While he supports any move to migrate 2D users to 3D as fast as possible, managing director of Nova Design David Parry is adamant that the ideal for any manufacturing user should be a single multi-functional commercial product. “Our view is that is essential to standardise on 3D and ultimately move to a single product; it makes no sense to have several packages doing the same thing,” he says. Nova Design started in 1988 offering engineering design services in the UK and internationally across a wide spectrum of disciplines, from heavy engineering, steel rolling mills and the like, to nuclear engineering, food processing and the motor and aerospace industries. “Our main deliverable from the CAD system is drawings,” says Parry. “We make massive specialised machines which need to be assembled and there is very little scope for automated assembly from CAD data.” Nova’s recognised very early on that 3D solid modelling was the only logical basis for its future designs. “We did our research, but the overwhelming criterion for us is the fact that Autodesk have such a huge user base and that DWG is such a standard format,” says Parry. “We don’t want to get involved with data translation or linking with other packages, because despite what people say data translation is not 100%.” Beyond that he observes that if you want to be in business in the future, “you have to do more for less, with less people. If I can’t draw something quicker today than I did a year ago the product is of no value. If you factor in the time taken to do translations and the costs, the product is not a commercial product – it has to accept native data and it has to be efficient.” Fair point. Which leaves us just needing to restate: your first thought should be the business needs, not the IT or CAD applications. Your second should by where’s the phone number or website for my 3D modelling upgrade. From then on, it’s about rethinking the links with the rest of the business as per your first thought, and building up with relevant alphabet soup.