PLM is about product, lifecycle, and management. Products are the raison d'être of manufacturing, so naturally the basis of PLM is integrated computer-aided design/manufacturing/engineering (CAD/CAM/CAE). Other applications manage and otherwise monitor other aspects of a product's lifecycle—those aspects beyond product design and manufacture, such as sales, service, and retirement. Last, PLM is all about the management of data and, equally important, the management of collaborations that go on throughout the supply chain. These are the interactions between people, business entities, production sites, and information technology (IT). "How do you manage all of these data and workflows into developing, manufacturing, and distributing products?" asks Ken Versprille, partner and PLM research director for Collaborative Product Development Associates, LLC (cpd-associates.com).
The answer: PLM.
That's the high-level view of things to know. Here's a handful of other things.
PLM is not a static concept, technology, or strategy. Once "just" a set of applica-tions on an engineer's desktop, PLM is a "strategic element of your business," says Ken Amann, director of research for CIMdata. "PLM is an evolving issue, just like your business is evolving. Is PLM enabled with technology? Absolutely. But it's as much about the process and how you do things as it is about what you build. While you can operate more efficiently with enterprise resource planning [ERP], and customer relationship management [CRM] lets you interact effectively with your customers, what makes a differ-ence to the bottom line is the products that you can sell and deliver and support. That's what PLM is all about."
Allusions to ERP are rampant when discussing PLM. Both started as several department-specific applications coalescing into an enterprise-wide business management system. However, explains Eric Sterling, senior vice president of marketing for Siemens PLM Software, PLM closes the loop between the information from manufacturing execution system (the AS-BUILT) and from CAD/CAM/CAE and ERP (the AS-PLANNED). Tracking and understanding this digital definition of AS-BUILT—what has actually happened on the plant floor—is becoming increasingly important in this age of government regulation, green design, safety, and liability.
Manufacturing's deep, dark, dirty secret for years has been that while somebody engineered something, what actually got built might be something else. Nobody really knew if the theoretical engineering information lined up with what actual product. Sterling says PLM provides the feedback loop that declares, "Okay, you didn't build what we engineered, but we are able to test what you built and tell you that it meets the [customer] requirements originally laid out." In short, PLM is the glue between upfront planning and everything else: detailed product design, manufacturing, and retirement. That glue becomes especially critical in the global environment, where, says Sterling, verbal collaboration is generally not possible. "You need an infrastructure to see change, manage change, and automate change across the globe." Or PLM.
These days, much of the focus in PLM is in additional functionality, specifically product and portfolio management, requirements management, business analysis, warranty management, and compliance management (green initiatives and government regulations). Coming to the fore, says Versprille, is the management of software revisions and CAE data. The hot buttons here include product design, test, reliability, and safety as they relate to life, death—and warranty exposure. A PLM strategy should include saving data from all product simulation trials. What's yet to be fully addressed is what CAE data to save.
As is so often the case in anything computer related, industry requirements drive software, which in turn drives the selection of compute infrastructure, IT implementation, and vendors. PLM is no different, starting with aligning an enterprise's business strategy with the PLM vendor's pedigree (engineering, financial, or information management) and industry niche.
Beyond that, PLM must be tailored by industry because the applications that matter most in an enterprise, beyond the core PLM applications, vary by industry. For instance, points out Al Bunshaft, managing director for Dassault Systèmes Americas (3ds.com), styling, designing, and safety are critical in automotive. Safety, product traceability, and product lifecycle tracking are critical in pharmaceutical. Pharmaceutical also uses styling and design applications, but only for designing the packages that appear on the shelves in stores. Component traceability is growing on automotive.
For companies surviving ERP implemen-tations, PLM can be daunting, even scary. However, points out Sterling, PLM is "a single unified set of domain solutions that work together. You don't have to buy them all. PLM can grow as your maturity level grows." Sterling urges "organic growth" for successful implementations. "You eat part of the `elephant,' gain success, and then grow from there."
Compared to PLM implementations even two years ago, today's PLM implementations are faster, data captures easier, and the workflows and business analytics are more relevant to the target enterprise. Out-of-the-box, template-based approaches provide fairly good "standard" PLM process flows. (Customization is still required through for "nuanced" business practices.) But, warns Amann, a "business issue that most people underestimate is the whole data-migration/data-cleansing problem—particularly when you're bringing information out of old systems. Frankly, a lot of that data is not very good."
The news is generally good, though. Plenty of examples exist of fast PLM deployments, companies up and running with PLM quickly, and speedy returns on investments. While industry-focused (a.k.a., prepackaged) PLM systems with built-in best practices (a.k.a., templates) have helped, more important is the Business Management 101 stuff: understand what you're trying to do, manage expectations, track metrics relevant to your business, prove you're benefiting and, if you're not, where to make corrections ASAP.
One basic aspect of enterprise manage-ment systems needs highlighting: "Enterprises that don't have a unified PLM strategy risk fragmentation, increased complexity, and much higher costs. Senior executive support is required to drive PLM. It's not something that an engineering department can stand up and substantiate," says Bunshaft. "This is where lots of PLM efforts break down."
Once upon a time, software required building an infrastructure of computers and data networks, loading in that software, testing and deploying the entire system, then running and maintaining the whole kit-and-caboodle. In the past decade, fewer major companies run their own IT infrastructure. They outsource it. In the last few years, many large companies have adopted two compute models for running critical and enterprise-wide software: software as a service (SaaS) and cloud computing. These are both pay-for-what-you-use models. The big stumbling block in both—and this applies to everything, not just PLM—is that data is off-site, outside the corporate firewall. "A lot of companies are afraid of that," says Versprille. That fear is changing. Increasing numbers of people and companies are backing up their computers to web-based services and are using web-based office programs, such as those hosted by Google and Microsoft.
"As smaller enterprises wish to collaborate and participate in larger product developments, the cloud may be a viable way to facilitate supply chain collaboration and coordination between OEMs and their distributed set of suppliers," says Bunshaft. "Also, cloud computing will become important as PLM starts to penetrate new industries. These new delivery methods make very sophisticated applications available to those who really can't afford, build, and maintain these applications themselves."
Also consider social computing. Here is "this notion of collaboration with much wider audiences—new types of communities being formed that let others outside traditional designers and engineers provide design input," continues Bunshaft. Facebook, Twitter, and similar technologies can be stimulating. They can motivate potential customers to experience products before those products exist. These people will provide feedback about those products. But social computing brings up new challenges, says Bunshaft. "Collaboration outside the traditional ecosystem of the OEM and its supply chain, is, frankly, something the rest of the PLM world hasn't gotten yet."