Steve Luby
President and CEO
Vistagy Inc.
Waltham, Mass.

Edited by Sherri Koucky

EnCapta software captures nongeometric information in   CAD models. This makes it possible to define surface treatments of a CAD-modeled   electronics enclosure.

EnCapta software captures nongeometric information in CAD models. This makes it possible to define surface treatments of a CAD-modeled electronics enclosure.


Information about five connectors in the electronics   enclosure can be easily edited in table view, manipulated, or sorted.

Information about five connectors in the electronics enclosure can be easily edited in table view, manipulated, or sorted.


Once the information is captured, it can be sent to   downstream applications such as Web pages and spreadsheets.

Once the information is captured, it can be sent to downstream applications such as Web pages and spreadsheets.


It sounds so easy. In fact, improving communication in product development requires careful decisions about the tasks people perform, the tools they use, and the information they share. Engineering managers must understand how these elements factor into the big picture to orchestrate the whole effort toward successful products.

Having a handle on existing design data isn't enough. Successful managers must identify information a team lacks and determine how to capture it. You can't manage or share important data you don't have. Also, make information available in a form that team members can easily use. Product design data that is difficult to access might as well not exist at all.

To understand the complex nature of design data, consider the tasks facing an engineer who is integrating the components of a laptop computer. An initial CAD model establishes the physical layout. The engineer sends out the model for finite-element analysis and receives back a report of potential flex points on the laptop shell. Now he knows where he can and can't mount a rigid printed-circuit board.

Next he must evaluate the thermal behavior of electrical components. So he sends the CAD model to a team member responsible for simulating heat flow with computational fluid-dynamics software. Back comes an e-mail stating that work can't begin without a list of internal components, their locations, and how much heat they generate. The engineer locates the component BOM, then spends a day searching for thermal data in spec sheets, catalogues, and handwritten notes from previous designs. He types that information in a lengthy e-mail and requests the heat flow simulation as soon as possible.

The heat flow results come back a week later. In the meantime, the engineer has finished a preliminary layout. As it turns out, the hard drive and the communication circuit must be moved out of a hot zone. Another day is lost redesigning.

Now the component layout is nearly complete. The engineer chats with a colleague from marketing, who mentions casually, "We're all excited about the snapin, snap-out installation for the hard drive. Our customers will love it."

The engineer never saw marketing's request. Fitting that feature inside the laptop will require a major redesign, including another round of thermal analysis. And the clock is ticking: The product must meet a development cycle two weeks shorter than the previous laptop.

As we've just seen, the complete definition of a product combines two kinds of data: geometric and nongeometric. A 3D part model, an assembly model, simulation results, and other such geometric data define the product shape and behavior. Everything else your team needs to know — in fact, most of the data about the product is not geometric. An enormous amount of documentation in product development records nongeometric information, such as bills of materials, process notes, material specifications, functional specifications, and engineering change orders.

All this data is vital to the engineering process, but it isn't necessarily available early or captured in a way that other people can use. Collaborative design requires feedback. Whenever important nongeometric design information is not digitally available at the time it's needed, the feedback loop breaks down.

A complete digital definition of a product captures both geometric and nongeometric design data. Making data digital and searchable is the key to making it accessible, so it can be exchanged between engineering, manufacturing, analysis, suppliers, and the business unit.

Getting to market quickly requires efficient information access and design practices. Less time searching for data and quicker release of design information lets engineers make informed decisions early in product development, when changes are least expensive and most effective at cutting manufacturing times and costs.

Another factor is the physical makeup of today's design teams. Collaborating is easy around a table in an engineering department. But to incorporate specialized knowledge and work around the clock, many organizations have widely distributed, even global, design teams that can include suppliers, manufacturers, and customers. Internet and intranet file transfers, e-mail, and the graphic capabilities of the Web encourage an unimpeded, and occasionally overwhelming, flow of design and manufacturing data.

Lack of compatibility adds another layer of complication to product data. Most engineering organizations use software from many vendors to meet their specialized needs, including multiple CAD systems and separate tools for analyzing stress, flow, tolerances, motion, and so on. Vital design information also resides in spreadsheets and programs for word processing and costing as well as in parts libraries in legacy databases. Most of these software tools do not exchange data easily, which means that communicating between them may involve time-consuming manual reentry.

Meeting each of these challenges requires organizations to collect product information in a digital, searchable, accessible, and exchangeable form and carefully manage that data.

Many product-development organizations use various design data-management systems and realize a degree of success at storing and retrieving important data. Data-management systems now must match the rapid progress made in design automation and address the continuing trend toward digitalization of all productdevelopment data, especially nongeometric manufacturing information from the manufacturing process disciplines.

Manual systems for managing design data usually involve a master folder for design documents and a central or distributed library for reference materials. Information resources include three-ring binders, handbooks, catalogs, and file drawers of memos and correspondence. Data searches are time consuming, requiring knowledge of the in-house filing system and the most likely place to look. To shorten search times, companies develop elaborate standards to ensure that the same type of information is stored in the same place. Frequently these standards migrate to digital design media without anyone questioning their continued use. Because data sharing involves constant manual re-entry of information, it's slow and not always reliable. Supporting distributed teams with a manual system is inefficient and expensive.

Product-data management (PDM) is a digital technology for vaulting, routing, and tracking documents such as drawings and bills of materials. PDM systems provide searchable metadata for finding specific information, for instance, part and revision numbers, the designer's name, and release date. Document check-in and checkout functions track workflow. Update notifications shorten the product-development cycle by informing members of a design team as soon as an updated document is available. PDM systems are typically used by engineering workgroups within a single company or division.

Enhanced PDM systems let engineers attach notes and memos to a file — for example, to explain the reason for a component change — and can extract limited amounts of data, such as weight, part number, and catalog number, from bills of materials created in CAD software. However, PDM systems generally treat CAD files as single data entities, not as searchable data resources. PDM is not a medium for driving up-front design requirements into a CAD session. Still, PDM systems promote communication and collaboration and have substantially shortened product-development cycles.

Design portals and project Web sites are intranet or Internet gateways through which collaborative teams can view and, in some cases, modify the various digital data sources necessary for product development. Project Web sites supply links to the same kinds of data available through PDM systems, but with a greater emphasis on supporting access to a broad audience and providing a kind of onestop shopping for product information. Potentially they also provide a customized context for information, which shortens search times. Portals and project Web sites also manage workflow and record design history, which encourages design and process reuse. However, portals and Web sites cannot guarantee that all required data exists in the system.

Collaborative product commerce (CPC) provides PDM-like functionality to the entire enterprise. It has additional interfaces for accessing data from computer-aided engineering, design and production, component and supplier management, design visualization, and enterprise resource planning applications, among others. CPC takes advantage of Internet and intranet technology to let each team member review existing design data and provide input. Nonetheless, the exchange of information in the early stages of the design process is still difficult. A design datamanagement system that is truly effective must have a means to collect and exchange all the data that defines a product. CAD software has evolved from being a pure geometry documentation tool to a 3D modeling environment that captures shape, behavior and, increasingly, design intent. However, it still does not allow the engineer to express all the important early design information presently embodied in engineering notebooks, spreadsheets, or other documents. And CAD information is not easily shared with other tools and team members.

Today, CAD vendors and their technology partners are providing more and more capabilities for storing and communicating data. For example, EnCapta software uses customizable templates to capture nongeometric data and link it associatively to the CAD geometry. The software lets design and manufacturing organizations store a complete digital product definition in their CAD model. It also helps product design by making it easy for members of a design team to exchange information early on, as features and mechanical behaviors are being modeled. Using eXtensible Markup Language (XML), the software can then make design and manufacturing information available to other applications such as those used by procurement, costing, and production departments. Such automated exchange of data shortens design cycles by reducing the time people spend searching for, reentering, and transferring information.

Software tools such as EnCapta enable organizations to implement an informed design feedback loop. The tools also promote collaboration by enabling PDM, CPC, and CAD systems to exploit the broad communications potential of XML. A number of PDM and CPC systems will offer XML compatibility within the next year.

When the time comes to implement a system for managing data, most organizations start by gathering under one roof all the information they already have. The next step is to decide what to do with it. This is why implementations often take two years and cost a fortune and still do not satisfy the informational needs of a development team. A swift, cost-effective implementation starts with understanding the needs of people, not with categorizing data. Discover who needs access to what information, and when they need it. Identify where the information comes from or should come from. Determine how to capture the information so it is accessible to the people who need it, when they need it. Implement your strategy for managing design data from the ground up, based on data needs and design tasks. Confirm that your system makes individual jobs easier and avoids data bottlenecks and knowledge gaps. Most importantly, make sure your system promotes collaboration early in design, which will pay off in better products and happier customers.