PLM Stratagies-Before You Tee Off or Suit Up30 Jun, 2006 By: Kenneth Wong
PLM advice from consumer goods manufacturers
If you trace back to the genesis of golf club maker PING (www.ping.com), you'll end up in the garage of Karsten Solheim, a General Electric mechanical engineer who decided to invent his own putter in 1959 after becoming fed up with those on the market. Five years ago, PING made a strategic decision to focus on PLM (product lifecycle management), PDM (product data management) and CAD for process improvement.
Figure 1. PING, a golf gear manufacturer, began using Pro/ENGINEER to produce iterations of its products.
"CAD allows us to do more iterations more often, so that allows us to improve quality and performance by exploring design alternatives," explains Dan Shoenhair, engineering director of PING. The company, which used to introduce two or three new lines a year, now introduces as many as 14 or so each year. Most makers of consumer goods and apparel face this challenge of getting multiple product lines to market in time for peak selling seasons.
A client of PTC (www.ptc.com), PING uses Pro/ENGINEER, Windchill ProjectLink and Windchill PDMLink for the development and production of its drivers, fairways, hybrids, putters, wedges and bags. Here are some tips Shoenhair shared for those considering a PLM implementation.
In this article
Flagsticks for PING and PTC
"I recommend a phase-in approach to PLM," Shoenhair says. "Give your employees the opportunity to understand the system, then build a consensus behind it. More than anything, have patience. Most companies aren't willing to do that—they want to put it in right away and get immediate benefit." In other words, the measured pace of a golfer is more advisable than the frantic tackles of a footballer when it comes to PLM. After each phase, Shoenhair advises, "Pause, take a breather, see if the software is performing correctly. If the users need more training, solve some of their problems, give them some attention before moving on."
"It's very important to not overwhelm people with automatically generated e-mails from the PLM system," Shoenhair warns. "[Employees] end up becoming desensitized to them. They'll dump them into the junk folder." It's a common mistake in first-time implementations, because management typically underestimates the threshold employees have for filtering their incoming e-mail traffic. He recommends instead a staged implementation or a trial period: "Starting from scratch, from having no concept whatsoever of PLM, we had a tremendous amount of initial training to help people see the overarching benefits of the system. We allowed our people to grow into it."
Figure 2. Since adopting PTC s PLM solutions, such as Windchill PDMLink, PING has dramatically increased the number of new products it introduces, from 2–3 a year to 14 a year.
"There's always resistance to change, to becoming more formal and organized, as opposed to free flowing and doing things as everyone personally sees fit to do," Shoenhair cautions. "So there has to be a cogent argument for conformity. Having a process that's well defined is not bureaucratic, though some people might say it is. The key is that if you define a process, it manages change better, more consistently. We document, we formalize, but the key is we keep changing—not for the sake of change itself but to improve. In fact, if the formalized process does not continually change, that's an indication something is wrong."
The Latest in Fashion
The sheer number of products and fast-paced development cycle in apparel and consumer goods companies puts a premium on quick access to information. Chris Groves, CEO of Centric Software (www.centricsoftware.com), remembers how he made one of his easiest software sales: "We were making a presentation to a fashion company. We were doing a demonstration of the decision dashboard [a feature of the software]." His audience included, in addition to the CEO of the company, an executive named Giorgio, who was responsible for providing the CEO with periodic updates. At some point during the presentation, the CEO took possession of the laptop and proceeded to explore the reporting tabs on his own. After a series of mouse-clicks, the CEO turned to Groves and asked, "You mean this is just like asking Giorgio what's going on?" As Groves recalled, it was an awkward moment in the boardroom as everyone took note of Giorgio's discomfort.
"The apparel and fashion industries have incredibly short product lifecycles," Groves says. "They have multiple seasons per year. They have inflexible go-to-market dates; they can't move Christmas. Typically, they'll do rounds of product briefs with a sketch or a description. Their sales people will try to obtain preliminary indication of demand while they're still designing, developing and sourcing. By a certain date, buyers make final commitments for purchase. The teams are almost always dispersed. Sourcing is done simultaneously with planning, development and sales. That's quite different from aerospace and automotive. It's not a system where one group hands off a product to another. We've seen people maintaining spreadsheets with 2,000 rows and 200 columns."
Groves observes that the volume of unstructured data involved—Word documents, Excel files and e-mails—makes it extremely difficult to answer a simple question like, "Are we on schedule?" He says, "Well, that depends on how each of the 50,000 items involved is doing and whether the top-selling items can meet the deadline. If not, why not? Is it because of approval delays? Is the preliminary demand big enough to justify putting in more people and effort? If you do that, will you lose money? Then should you just cut your loss and drop the items? What is the financial impact? To get a full picture of the situation, they need to get database reports from ERP or PLM systems and roll them up. More often then not, they don't have those systems. Even if they do, they can't lay their hands on the information quickly."
If you're in the fashion and consumer goods manufacturing business, Groves predicts that you likely belong to one of the two types of enterprises: "On one extreme, you have 90% of the information in some managed system: SAP, TeamCenter, Microsoft Project and so on. On another extreme, you may be tracking everything in Excel." In either case, he recommends that you take stock of the information you need to make critical decisions: "Ask yourself—if you could have access to all the reports, what information would you like to see? Get an inventory of the current reports or Excel sheets. It's a good starting point."
Centric recently announced its Collection Management solution targeted at consumer goods and apparel manufacturers. With emphasis on connecting disparate applications rather than consolidating them, Centric offers product intelligence solutions for collecting and presenting live business and product data in Web-based decision-making dashboards. Centric's technology can search and index structured and unstructured data automatically. The system, Groves explained, can manage either copies of documents or references to documents, so if you're not prepared to consolidate all your enterprise data physically in a single repository, you don't have to.
Under the Circumstances
There's no shortage of PLM vendors with strong opinions about why they think manufacturers should adopt PLM products, preferably their products. So to instill a measure of frankness in our discussions, I recently began asking PLM vendors to address what might be their least favorite topic—the circumstances under which they might advise someone against PLM.
Tom Shoemaker, PTC's product marketing director for Windchill, observes, "There can certainly be cases where fixing the PLM system might not be a potential client's top priority. That might be the case if they've got a far more pressing issue, like managing accounts receivable or stock inventory—something PLM won't solve directly. If a company feels like it's operating as effectively as it can, and it's getting high-quality products to market as quickly as it should, despite global competition, then PLM might not be a priority." On the other hand, he notes, there aren't too many companies that are wholly satisfied with their performances in those respects.
Centric's Groves predicts, "Companies with different divisions or multiple design and manufacturing sites—design anywhere, build anywhere—using disparate PDM and CAD systems will have a difficult time both technically and politically standardizing and centralizing on one PLM system. Each division or location will want to continue to use familiar applications and processes. Furthermore, companies with offshore manufacturing may not want to bring their offshore partners into the PLM application for security purposes, to avoid automatic e-mail notifications, for example." If you belong to this camp, he suggests that you consider other alternatives. One would be consolidating all of your critical enterprise and product data using a connectivity solution that combines data from multiple systems, like Centric's.
Kenneth Wong is a former editor of Cadence magazine. He explores innovative use of technology as a freelance writer. E-mail him at firstname.lastname@example.org.