Lean Six Sigma helps Xerox engineers achieve faster time to market
By Mark Enzien
Maintaining rated speed on heavier paper stocks is old news for the most advanced xerographic digital color presses, such as the Xerox iGen3® 110 Digital Production Press. But in the slightly less expensive and less productive class of digital color presses that emerged in this decade, production typically slows on heavier stocks. While these presses have performed well, customers increasingly ask for models in this class to retain rated speed no matter what stock they run.
To meet this need, Xerox engineers last year made an internal proposal to retrofit an existing product in the category, the Xerox DocuColor 8000 Digital Press --and they were promptly challenged. The development team that developed the press thought the retrofit wouldn't work. Even supporters thought the development schedule was overly aggressive -- and that delivery would take twice as long as predicted. Time was a factor in another way as well. The engineers developing the solution were based in Webster, N.Y., while product design and manufacturing occurred in half way around the world at Fuji-Xerox in Japan.
Even supporters thought the development schedule was overly aggressive -- and that delivery would take twice as long as predicted.
These are just a few of the obstacles Xerox engineers overcame in delivering the new Xerox DocuColor 8000AP Digital Press in what the team believes is record time. The new press becomes available this week, less than nine months after the development program began.
Turning Up the Heat
The DocuColor 8000 prints at speeds as fast as 80 pages per minute (ppm), but slows by as much as half on heavier stocks. The reason: heavy weight stocks require more fusing energy (heat) to ensure that toner attaches firmly to the target substrate. To generate sufficient heat for the more demanding work, the system has to slow down.
The iGen3 development team had solved this problem by using external rolls and new temperature-control algorithms to add heat to the fusing system. The new Xerox DocuColor 8000 fuser system design team believed the concept could be applied to its project. However, the Fuji-Xerox design team members weren't yet convinced. So, the two design teams met in a workshop to consider the merits of the approach. By the end of the meeting, the entire team --especially the Fuji-Xerox members-- championed the idea.
In December 2006, With new development tools and efficiencies in place, the engineering team believed it could complete the project in much less time; by the end of 2007 or beginning of 2008. But that wasn't fast enough, according to management, which called for the new product's completion by July 2007.
A team of scientists, engineers and marketers decided to pursue the project. Then came the bad news: Similar previous projects had taken about two years.
With varying degrees of enthusiasm for the program timetable, a team with 10 core members and 30 extended members signed on. Its leaders included Technical Program Manager Bob Mara, Lead Fusing Engineer Steve Russell, Program Manager Tom Bitter and Don Bott. They were the prime drivers in taking the project from concept to the physical demonstration.
The work began with strong evidence that the external roller approach would work, based upon modeling simulations performed by the Xerox design team. The modeling simulations also gave the team a head start in developing a hardware prototype.
Two other elements of the development program worked in the favor of accelerating TTM (time to market). One is that the project focused on a single performance metric in a single physical system. The team could attack it full on. The other is that the team used DFLSS --Design for Lean Six Sigma-- tools. Lean Six Sigma is a methodology for achieving the fastest rate of improvement in customer satisfaction, cost, quality, process speed and invested capital. Xerox champions this process by selecting the most appropriate Lean Six Sigma tools and by providing rigorous staff training, and Xerox engineering has embraced the program. Of the 70 Xerox Production Systems Group staff members who have achieved black belt status signifying the highest level of accomplishment, 62 are in the Platform Development Unit, which also generated more patents than any other Xerox division this past year, 112. Most members of the fuser development team had Lean Six Sigma training.
For this project, the team worked with software from Air Academy that interprets Lean Six Sigma concepts to guide the designs of our experiment and dramatically reduce the time required to narrow the field of possible solutions. It eliminated the need to try every possible combination of factors, so the team could run fewer physical tests.
The team's focus was on the temperature of the fusing system, which is controlled by software and influenced by the hardware, thermodynamics, and the timing of components contacting each other. The experiments worked like this: Each time the team changed a heat control algorithm, the Air Academy software applied a process to quickly adjust all the other fuser software settings for optimal performance. Then the team ran the experiment, generating date that is analyzed with other DFLSS software tools.
Software from Air Academy interpreted Lean Six Sigma concepts to eliminate the need to try every possible combination of factors so the team could run fewer physical tests.
From that, that team developed a follow-up experiment in an iterative process that led to the optimal solution.
Research On a Roll
By taking the project one step at a time --and completing each step on schedule-- the project leaders gradually led team members to recognize that the program was on track. Skepticism began to fade.
Another challenge was overcome with cooperation from Fuji-Xerox. No one on the development team had experience with the product line, but their Fuji Xerox counterparts helped them come up the learning curve. Cooperation was critical, because even as U.S. engineers developed the software and machine changes, Fuji Xerox continues to manufacture it.
The team overcame many other challenges by thinking simply to help avoid more complex paths to a solution. For example, one reason DocuColor 8000 images were slow to fuse was the high-gloss finish, which requires more toner and therefore, more fusing energy. The new system provides a less glossy finish, which requires less toner and fusing energy and is closer to what most customers want. That noted, the new solution also provides an option of printing with the same high-gloss finishes the press has been known for.
In May 2007, Xerox announced the new Xerox DocuColor 8000AP Digital Press, which retains rated speed on all stocks. This week the new press becomes available. The techniques that achieved the rapid-time-to-market while maintaining a close-to-normal work schedule for the staff are being applied across the board to improve TTM in many projects. In addition, the development team is now investigating extending the upgrade to other products in the DocuColor 8000 family.
And next time a project comes up that requires a faster turnaround than has been historically possible, this program will remind Xerox engineers of what is possible with Lean Six Sigma tools and a dedicated team.
Mark Enzien is vice president, Platform Development Unit, Production Systems Group, Xerox Corporation. He can be reached at [email protected].