An Introduction to Virtual Product Modeling
By Nick Sale, INCAT

Design data (information created during a product's life cycle) together with design know-how (the knowledge that designers accumulate over time) are increasingly recognized as critical business assets. When leveraged properly, they can reduce lead times and increase quality dramatically. As one magazine writer recently observed, however, trying to identify and manage these so-called knowledge assets is like trying to fish bare-handed. Having presided over numerous product data management (PDM) implementations in recent years, I know all too well what that writer meant.

Nowadays, thankfully, software tools make the process a whole lot easier, bringing the benefits of PDM within reach of just about every manufacturer. Easier does not always mean clearer, however, as those of us in the PDM arena sometimes forget just how confusing it can appear to those new to it. In the CATIA community, especially, this confusion is compounded by an increasing array of technical terminology neatly condensed into groupings of three or four letters: CDM, VPM, PDMII, VPDM and so on. This article, therefore, is an attempt to demystify PDM, and to explain its connection to one of those three letter groups in particular-VPM or, to give it its full name, ENOVIAvpm. ENOVIA Corp. is a subsidiary of Dassault Systemes, created in partnership with IBM in 1998. The VPM part of the name is an abbreviation for Virtual Product Modeling. By the time I have finished discussing what it can and cannot do, I hope to have dispelled the three common misconceptions surrounding VPM, namely:

1. It's only relevant to large OEMs in automotive and aerospace industries.
2. It's very complex and costly to implement.
3. It needs advanced CATIA users to produce real business benefits.

Let's start with the basic concept behind PDM.

PDM-some background
PDM systems were developed originally to capture, store and control all product-related information. This usually included design geometry, engineering drawings, part files, assembly diagrams, product specifications, project plans, machine tool programs, analysis results, correspondence, bills of material and many others. PDM's overall aim is to manage product data throughout the enterprise, ensuring that the right information is available to the right person at the right time and in the right form.

It is therefore an integrating technology connecting many different areas. PDM improves communication and co-operation between geographically and functionally diverse groups in an organization. It forms a platform to restructure product development processes and to implement time- and cost-reducing initiatives, such as concurrent engineering and collaborative product development.

What is missing from most PDM systems is the tight coupling with design data. For example, the Bill of Materials (BOM) in a typical PDM system is not connected to the assembly design. As a result, creating a BOM represents duplication of work and an error-prone design activity. Likewise, during the concept and development phases of any new product, the controls within conventional PDM system could not keep pace with the fast-evolving work of the design department. The restraints can lead to creativity and innovation being stifled. Limitations such as these led to the vision of a next-generation of PDM systems, generically named PDMII (see Figure 1).

The products from Dassault Systemes that realize the PDMII vision are ENOVIA LCA for larger companies and SMARTEAM, an NT-only solution for small and medium sized companies.

Where VPM fits in
CATIA Data Manager (CDM) is used essentially for vaulting of CATIA data and assembly management, and was one of the first tools that enabled digital mock-up. ENOVIAVPM takes the principles of CDM significantly further toward PDMII. It does this by incorporating PDM functions such as basic workflow, data security and configuration. This allows roles and responsibilities to be defined; it also permits creation and change processes to be built into the data management environment.

So now, not only are we able to provide the right information to the right people in the right form and at the right time, but this time from the very beginning of the product development process. This is an important difference because it allows design knowledge generated during the innovation phase of product development to be captured-something those traditional PDM systems were never capable of doing.

What VPM does
At the heart of any data management system, VPM included, is a database that acts as the vault for all the data and knowledge being captured. Specifically, VPM provides a single accessible environment where all product data (CAD models, FEA, test data, etc.) can be captured and reused from the start of a new product development process, without imposing constraints. Most designers spend anything between ten and 40 percent of their working day just searching for information. One of the simplest justifications for a VPM implementation is its ability to reduce this. It's also commonplace for designers to find themselves working on the wrong version of a model-or perhaps an earlier release that has been superseded without their knowledge. VPM prevents this by imposing restrictions on who can make changes. When a model change is authorized, it allows the impact of that change to be assessed in advance and, when made, propagates it to all related models, cutting ineffective design time further still.

Parts rarely stay at Revision 1. As parts are modified, many companies create a new model with a modified name to reflect the new revision. After a number of iterations, the current product is the sum of the latest revisions. The problem with this approach is that the only configuration defined is the latest one. It's very difficult to go back to view a snapshot of the product at an earlier date, or define a configuration for a different market. With VPM, as you create a new revision you also define the rules assigned to it-for example "from date x" or "with the French version." By applying these rules, the specific product structure required is generated.

So much for what VPM can do, what are its limitations?
VPM is not intended to do the enterprise-wide job of a traditional PDM system-it is focused on design data. Nor will it manage CAD data from multiple systems as effectively as it manages CATIA data. VPM is not an out-of-the-box solution, though it can be implemented rapidly. Effective data management solutions are closely related to the efficiency of a company's internal processes. Therefore, these need to be reviewed, and, if necessary, improved before substantial benefits can be realized.

Who might use it and why
VPM is often thought of as only for the big guys-the Boeings and DaimlerChryslers of this world. It's not. It could make sense for any CATIA user, depending on the data management challenges. Naturally, some implementations can get quite complex. Once defined, however, VPM is really quite straightforward to use and you don't have to be a high-power user to start seeing quick returns: novices can easily be up to speed within two days.

Cost and protracted implementation periods aren't necessarily prohibitions either. Ten-seat installation starts at around $80,000 (includes consultancy time, training and all software costs). And, by using an experienced consultants to expedite implementation, users can be up and running in around ten working days. If your designers earn $60 per hour, a time saving of three hours a week for each designer would justify the ten-seat example in around 44 weeks. (80000 divided by 1800 (60x3x10) = 44.4) This justification is before any other benefits are taken into account.

Bosch Braking Systems, based in South Bend, Indiana implemented VPM in the fall of 1999. Larry Delp, director of engineering technical services at Bosch Braking Systems, claims to have reduced product development times by 30 percent and dramatically shaved time off engineering changes.

Bentley Motors Cars in Crewe, England, a 70 seat VPM/CATIA site, has been using VPM for more than a year. It now manages multiple projects and configurations via its VPM Database. John Unsworth, Vehicle Packaging and Configuration Manager, points out that prior to VPM, it was quite a challenge to trace the progress of a design and mature it with respect to multiple vehicle configurations. Now though, full control and traceability of parts as they evolve via the DMU process is possible. The company has always been proud of its ability to offer customers a wide range of options, though of course these must be catered for during the concept stages of design. The new process, supported by the tools offered inside VPM gives the company the ability to do just that. "Eventually", adds John Unsworth, "we are aiming to link our VPM-driven DMU to give prospective customers a digital view of their chosen vehicle, complete with all the options that they selected."

ENOVIAvpm and SMARTEAM
Although very different products, choosing between them is very simple. VPM only runs on UNIX (although a limited functionality is available on NT via ENOVIA 3d COM) and manages CATIA Version 4 and CATIA Version 5 data. SMARTEAM only runs on NT and manages CATIA V5 data only. So if you want to run V4 and V5 with a single database, your only option is VPM and UNIX for both versions of CATIA. If you want to be 100 percent V5, then NT and SMARTEAM is an option.

Conclusion
Well-organized product data can help any size organization achieve advanced levels of design efficiency. It is also a prerequisite for progress to the next level of sophisticated inter-company communication: a lofty plateau where an OEM's design changes are rapidly propagated throughout its supply-chain, leading to even greater reductions in cycle-times, inefficiency and cost. The goal is to enable a product's entire evolution to be captured-the design intent as well as all the actions that created it and the multitude of inter-relationships that exists between parts, components and assemblies.

Returning to the fishing analogy, with tools such as SMARTEAM and ENOVIAVPM, our fisherman has a net at his disposal. All he needs now is to be taught how to fish effectively.

About the author

Nick Sale is Head of Consulting Services at INCAT Ltd. in the UK. Based at INCAT's North American operation from 1997 to 1999, he had extensive experience in defining and managing CATIA Data Management implementation projects in large, complex enterprises, including Boeing RocketDyne (makers of the NASA Space Station) and TRW-AEG (auto electronics). He was also consulting operations manager on the VPM roll-out at Bosch Braking Systems. Nick can be reached by telephone at +44 1582 878753 (direct) or +44 7831 519772 (cell phone), email nsale@uk.incat.com.

Generic Abbreviations:

VPDM - Virtual Product Development Management
PDM - Product Data Management
PDMII - Next generation PDM (PDM + VPDM)

Products:

CDM - Dassault Systemes' - CATIA Data Manager (Early VPDM system)
VPM - Dassault Systemes' - Virtual Product Modeller (VPDM system)
ENOVIAVPM - New name for VPM from ENOVIA
ENOVIA LCA - New product that should eventually replace VPM with full PDM capabilities