Lean Beyond Production
There is a Lean Thinker in all of us. Who hasn’t fidgeted in a grocery check-out line, endlessly awaited approval on a business application, or sat through a rambling executive meeting, frustrated by the lack of process, the waste of time and energy, and the glaring inefficiencies. Today, Lean Thinkers are easing such customer angst, identifying value streams in workplaces and business processes where many still see only random action, eliminating waste and increasing value-creation where others see only work.
Although challenges exist as lean moves further away from recognizable products and production processes and steps and transactions become less defined, lean principles now are being applied to a range of business processes. Lean principles are intended to reduce labor, space, capital, and time in delivering the right products or services to end customers. Using the same lean tools that worked so well in production, Lean Thinkers now are capturing improvements and benefits off the plant floor, in environments as diverse as an insurance company, academia, or the front offices of the world’s largest automaker.
“The light bulb goes off in people’s heads when they realize that waste is everywhere in any business process,” says Bob Emiliani, associate professor in the School of Technology, Central Connecticut State University. “It doesn’t have to be operations—they can look and see HR processes and financial processes and the connections between functions. And they ask, ‘Why does it take so long to promote somebody or hire somebody or close the books at the end of the month or quarter?’ They start to see waste everywhere, and once that happens people start to see it as not just a manufacturing thing.”
Amid the dogma of academia, Emiliani is challenging tradition through his work with lean. In the past, Emiliani had applied lean principles at companies such as Pratt & Whitney, working with supply-management functions. When Emiliani entered academia, he naturally brought lean principles with him and began applying them to how he taught his graduate course, “Leadership and Organizational Improvement,” when at Rensselaer Polytechnic Institute. While there he also led a kaizen review at Rensselaer to improve courses in an executive graduate degree program.
Although the environment certainly changes as lean moves beyond production, the tools remain the same. At General Motors Corp., the lean improvement approach extends across the company—engineering, design, planning, finance, HR, purchasing, vehicle sales, service, marketing, etc.—and is incredibly similar to the tack taken with production, says Lou Farinola, director of the General Motors, Global Manufacturing System (GMS), Integration Center. In fact, GM tackles its lean improvements in non-manufacturing areas explicitly using its “manufacturing system,” the GM equivalent of the Toyota Production System.
“One of the things that was good and a credit to our leadership was that everyone said, ‘We are a manufacturing company,’ and if we think about us as a manufacturing company, the concept of applying the Global Manufacturing System in other functions shouldn’t bother anybody,” asserts Farinola.
The foundation elements of GMS in the office consist of business deployment, process flow (level scheduling), in-process control and verification (andon), standardized work, workplace organization, and people involvement. There was some concern that people would resist the elements because of “manufacturing stuff getting crammed down their throats,” says Farinola, but for several reasons that hasn’t occurred: GM leadership openly supports the use of GMS in manufacturing and results there have been impressive; it’s been thoroughly communicated that the same concept that worked in manufacturing can be applied to non-manufacturing; and, most importantly, improvements that take place due to GMS in non-manufacturing areas speak for themselves. “When people see results, they’re less inclined to worry about the name of it and are willing to get on board.”
Seeing the process
While the movement of lean into non-production business processes is underway, it’s far from ubiquitous. For example, 55% of manufacturing plants indicate the primary improvement methodology they follow is lean manufacturing, lean and six sigma, or the Toyota Production System, according to the IndustryWeek/Manufacturing Performance Institute 2004 Census of Manufacturers. (That percentage seems large, but 37% of the lean plants report only “some” degree of implementation, which conceivably could be as limited as 5s.) Of all plants along the lean journey, a fraction has expanded it to other business functions: purchasing (at 37% of plants), engineering (28%), finance and accounting (15%), supplier relations (18%), customer relations (18%), administration (18%), and R&D (8%).
The numbers hint at the challenges of applying lean in non-production environments—lack of process orientation, inability to identify the customer, little standardization, and a propensity for rework.
“All of the office work is more difficult,” says GM’s Farinola. “There is no question that there is difficulty because of so many factors that make office work office work.” One of those factors is often an inability or unwillingness of staff to see their work as a process or within a process. When GM expanded GMS to other areas of the company, value-stream mapping was a critical tool to get over this process-visibility hurdle. Unlike GM’s vehicle assembly areas, where value-stream mapping was less likely to be used because processes and steps were so physically apparent, maps in the office make processes tangible.
“In assembly, it’s much easier to see how material and information flows, the steps for doing the operation, and how steps relate,” says Farinola. “All that was very visible and could be studied, analyzed, and improved upon. In the office, the process is very invisible. So the hypothesis was that if we can somehow make visual the work done in an office, we can do GMS just as we do it in an assembly plant. We did a pilot project in finance, where we demonstrated that the business processes in the office are not very different than in manufacturing, it’s just that you can’t see them—and therefore we don’t study it, measure it, and improve it.”
Additionally, knowing how good or bad a business process is can be difficult because many processes and steps simply are not measured or not measured accurately or correctly: for example, call centers may track calls handled but not permanent problem resolutions; finance departments track staffing required to close books but not timeliness of book closings; or service establishments track customer volume but not complaints per customer served.
Just as in production, adds Farinola, employees will “monkey” with the numbers to look good. GM addresses that by developing metrics that more accurately reflect what needs to be measured and tries to get staff to be honest about the data. “If you set up a system where people are going to get beat up for being in the red, they’re not going to be red,” he warns. “And so you don’t see the problems, you don’t work on them, and you have no way of calling for help, which gets you to andon.” For example, if an engineer doesn’t have information (i.e., material), he can’t work. “How do we visualize that, how do we track that, how do we improve that, and what do we put in place to provide support for that?”
In 2000, senior management at Jefferson Pilot Financial (JPF) appointed a newly established lean team to transaction-oriented areas where the ability to impact processes was most obvious and a need to improve service urgent. The successful initial efforts, which advanced from a model cell across the company’s new-business unit, were publicized by the Harvard Business Review in October 2003. Manufacturing tools such as takt time, level scheduling, and redefining metrics transformed the service unit and set the stage for a wider lean implementation that is moving across the life insurance and annuities company.
Since the initial rollout, the JPF lean team has moved on to functions such as licensing and activation of agents, underwriting, customer service, annuities, and product development. As lean progresses from clerical to more professional activities, workforce creativity often gets confused with value-creation, says Cynthia Swank, VP Lean Manufacturing and Strategic Project Office. “Professional functions are less process-oriented; they’re not like transaction processes.” The application of lean becomes more complex in such settings, she says, but the opportunities for improvement are just as tangible.
Recognizing the Customer
A tenet of lean is to establish processes focused only on adding value for the customer. Simple enough, but it’s tough to develop those processes when you don’t know who the customer really is. For Emiliani, understanding his customer was a significant step in applying lean to course instruction.
“Who are these customers, why are they different, and how would we do things different if we’re going to recognize and respond to the customers’ needs?” asks Emiliani. “That’s actually somewhat controversial in an academic setting to look at students as customers.” The “customers” he identified for the executive graduate program are part-time working professionals; they have jobs, families, pressures, and time constraints unlike a traditional student. “Most of the controversy stems from the fact that they are students and they are customers—it’s not an either or situation. But ultimately, if you understand these wants and needs better, you have more satisfied customers that speak well of the organization.”
Lean principles underscore Emiliani’s interaction with his students/customers, and they like the approach [see sidebar article Lean in Academia]. Nonetheless, other “operators” (i.e., faculty) have not necessarily picked up the approach. “They’re steeped in tradition, and there may be local customs that dictate why they do things the way they do,” Emiliani says. “One of the things about kaizen is that it’s a process of critical thinking that says, ‘Why do we do it that way?’”
Emiliani says a failure to view students as customers may result in faculty or administration not fully appreciating student feedback, and that can impact a student’s perspective of the value they receive. For instance, lean can be applied to a gamut of student experiences—registration, grievance processes, plans of study—all of which can adversely affect student/customer satisfaction with the institution, even if the classroom experience is good.
Chris Rice, Assistant VP Lean Manufacturing, says that every value-stream mapping session at JPF begins with the questions, “What is the value of this transaction? Who is the beneficiary of this, and how does it add value to our customer?” By taking this approach, JPF’s customer service unit was able to analyze 450 distinct process transactions (e.g., beneficiary change, premium receipt) and reduce that to 250, combining some steps and cutting others completely. “People had been doing them because they’d been doing them,” says Rice of some activities.
Understanding and delivering what the customer values is essential, but providing that in a service environment as “customization” can be difficult. “More and more people will be facing the ‘Burger King have it your way’ approach,” warns Swank. “We’re doing business with large groups of agents, and they have a certain way they want to do business with us, and they demand we do business their way. So it’s a real challenge for us how we manage that and still have a competitive advantage in terms of being a low-cost provider.”
Swank says JPF addresses customization in several ways, including use of “separate production lines” for custom business. Lean also has helped absorb customizations into processes without noteworthy increases in staff, she says, but there is a complexity unique to service companies: Unlike in the auto industry where a buyer may be willing to wait three weeks for a customized vehicle, service industries have no such luxury—customers want customization now.
Similarly, establishing close customer relationships appeals to many service customers and is seen as real value, but it, too, can throw lean a curve. For example, load balancing at JPF establishes a framework where agents are assigned to work with a group of underwriters, but many agents prefer one or two specific underwriters based on past relationships—and thus a backlog can form at favorite underwriters. Unlike in manufacturing where work can be balanced between cells or machines without thought to relationships, in a service setting relationships can exist (and often should exist), but this can introduce variability into work distribution, and causing variability typically is not considered lean.
Standardization vs. Creativity
Standardization helps to reduce variability and eliminate waste, but standardization is rare in most office environments. Just as on the production floor, the initial lean application of standardization in the office is often viewed by employees as regimental and hampering creativity. Farinola says that most professional workers perceive that everything they do is value-added, and so the whole concept of standardization and waste—while not foreign to them—doesn’t apply to what they do.
Farinola uses two analogies to describe the benefits of standardization with regard to perceived creative work: Artists will set up paints and tools in a consistent manner so that they can focus on the creative component—not reinvent every step once they approach their canvas and palette. Similarly, the military is synonymous with standardization. Yet fighting a battle is dangerous creative work where routines can easily break down—standardization enables soldiers to stay focused on assignments and perform rote tasks amid chaotic conditions.
“Probably the biggest challenge that we’ve got today with lean in the more professional areas is that professionals derive a great deal of their value from being able to work or work around the system—creative solutions,” says Carol Sineath, JPF Lean Manager. “That’s pretty unlean, and we’re challenging people to come up with a standard way to do their work. We sense that they feel we’re taking away some of the value in what they do.”
“They say they were ‘solving problems,’ but all they were really doing was eliminating symptoms of problems. It was a classic chaotic environment,” adds Rice. “In essence what we’re saying is, ‘You can do your job better, and we’d like you to do it in a standardized way.’”
JPF is applying standardization to its underwriting process and using an electronic work-distribution model there to cut waste. Underwriters accept some standardization for the end-to-end value stream and process handoffs, says Rice, but they’re not comfortable standardizing the creative actions of how they actually evaluate policies and assess risk. Some are even hesitant to standardize how they take notes—although many reassess and redo notes because they’re not comfortable with initial documentation, even their own.
In many business processes, idleness is one of the few activities that everyone inherently sees as waste. So it’s not unusual for staff “to get a jump” on projects, frequently moving forward with information that’s in a fluid or unofficial state. While the effort is admirable, the outcome is frequently waste. Farinola says he’s seen evidence of this since GM’s pilot project and on many projects since.
“People think they’re doing a good thing by getting a jump start as information becomes available,” says Farinola. In product development, for example, the engineering department may release information early, which is subsequently picked up by purchasing to get pricing, industrial engineering to develop layouts, and many others, and “they all believe they’re creating value by starting early. What we find over and over is that all they’re doing is generating the waste of correction. All this stuff will change, and it’s very predictable that it will change, and it will maybe change several times before you have the information to do the job right.” With GMS, staff are now identifying processes that they should not start until good data is in hand and they can perform their role right the first time.
Similarly, JPF is moving toward a Toyota-like approach with product development. The company uses product champions and gets all participants to recognize a development process and how they impact the process. All constituents get involved up front and address critical issues before resources are exhausted. Swank says that in the past customer service might have said, “We don’t need to be involved until the policy is rolled out,” but when rollout occurs they would declare that they can’t service the policy.
“Having everyone involved often shows that if you do one or two things differently, you can streamline the whole process,” says Swank. Because product development now is more efficient and has an improved track record, the group gets upfront approval for products based on the concept and business analysis of customer need and has exceeded the company goal for product introductions and enhancements.
Best Practices and Benefits
Business processes adapting lean would do well to consider the best practices recited by GM’s Farinola: “A leader who wants to move into the area of lean ought to first get a little knowledge about it. They should read Learning to See and Lean Thinking, and those kinds of things. They should get themselves a sensei—if they don’t already have one inside—to help them get started. Then they should get personally involved in doing a project, even if it’s a small one; engage themselves and their staffs and prove to themselves that this can really work. Before going gung-ho around the whole company, do a project or two at the leadership level and verify that they can see how this thing works, then they can talk the talk and lead by example.”
That approach has enabled GMS to be particularly effective in the office at improving quality, responsiveness, and cost, says Farinola. Prior to GMS, many processes reported first-time quality as low as single-digits (approximately 5% to 15%); after just the initial workshop, first-time quality typically rises to 60% to 90%. “Most of that comes from [improvements] where people have been working on insufficient information or have no clear understanding of what the end result needs to be for the customer,” says Farinola. In addition, lead times have been cut by half, and GM has counted “tens if not hundreds of millions of dollars we’ve been able to reduce.”
Cost reductions at GM have come about by establishing lean processes with external contractors—expenditures on parts, ad agencies, printing services, etc.—as well as internally by reducing work content. “In areas where we targeted for structural cost reductions and the majority of the cost [was expected to be] people cost, we’ve been able to work the attrition curve and not backfill people,” notes Farinola. “We’ve established targets, and [GMS] efforts have allowed us to meet those targets without trying to put the same amount of work on fewer people.”
At JPF, lean work in licensing and activation slashed lead times and improved customer service. Significant backlogs existed prior to lean, but by implementing pull processes and redefining the process around value add, the backlog has disappeared and the turnaround time for activating an agent has been cut by approximately 90%. For example, instead of spending significant time and effort to appoint an agent that may or may not supply business, JPF now waits until the agent sends in business (provided the agent is working in a state that permits this). Now, while a policy from a new agent works through underwriting, the agent gets appointed.
Prior to lean, non-value work at JPF included one group to remove non-producing agents (many of which should not have been appointed in the first place), and another group was dedicated to making calls to expedite appointing work. “Once lean was all said and done, those calls disappeared, and at the end of the day they stopped ‘servicing’ and started ‘producing,’” says JPF’s Rice. “All the noise went away, all the complaints went away, the phones came down to a standard level of Q&A, and now those are calls add value to the process. It was fascinating to watch.”
Sineath adds, “We’ve had a tremendous growth in the number of new business applications we’ve had over the last few years, and having these lean initiatives where head count was freed up has enabled us to absorb the volume increase.” In other areas, JPF has been able to reduce human resources (principally through attrition), and it’s been able to reassign staff to emerging businesses, such as the brokerage market, without having to add resources.
Another facet of JPF’s lean effort that promises exponential benefits across the company has been using lean as a foundation for putting in new technologies. Rather than deploying information systems ahead of or concurrently with lean, business processes are first analyzed, stabilized, and streamlined before any automation and IT is added. “We’ve put in a new workflow and imaging concept that deploys lean concepts, and we standardized the work of two different locations,” says Swank. “Since it’s all electronic, it’s seamless. We can ship work between locations. That’s where technology and lean meet and hit the road. Had we been non-lean, we would have done one location and then we would have done the next location, and we never would have had the synergy of being able to move work between these two areas.”
For Emiliani, his lean results often show up as satisfied students/customers, as evidenced by satisfaction scores with the course and with the professor. “As I’ve gotten better incorporating lean principals and practices and figuring out how they apply to my world and with the student experience, you can see the numbers improving,” he says.
Emiliani also will post anonymous student comments about the course with his syllabus, allowing future customers to see the value they’ll be receiving. He says, “Students get the sense that you’re serious about continuous improvement, that it’s not just a bunch of hot air. And that you’re serious about respect for people because you understand who these people are and what issues they have in life, and the ways they could learn better compared to traditional methods.”
Lean in Academia
Bob Emiliani, while the clinical professor at the Lally School of Management and Technology, Rensselaer Polytechnic Institute, began applying lean to academic work. He structured a graduate course there with lean, keeping two Toyota Production System pillars in mind—continuous improvement and respect for people. In doing so, he applied a range of lean principles common in manufacturing:
· Identify the customer: Emiliani recognized that working students (customers) in his executive-level management class were far different from traditional students and, thus, had different needs and wants from the class and different constraints on their capacity to succeed.
· Customer value: A concise, visual class syllabus helped students understand the value they would receive from the course. Emiliani also solicited anonymous feedback midway through the course so student comments could influence the balance of the course and he could incorporate those changes for the subsequent semester (while he waited for more formal feedback from university). He also posted anonymous student comments with the syllabus so prospective students could see what others think of the course.
· Eliminate waste: In graduate courses there is traditionally a significant amount of reading, but Emiliani worked to hone reading and keep it thematically consistent with course value. “Most professors assign big papers that they don’t want to read and students—especially working students—don’t want to write. Are you really getting learning from that? My standard format is to present things in tables and summary fashion and incorporate cause-and-effect analysis.”
· Root-cause analysis: Problem-solving tools such as the 5 Whys and fishbone diagrams often were used in the class to help students understand the root cause of business problems that relate to leadership and to find countermeasures.
· Scientific method: Emiliani says business-school assignments tend to be fairly ambiguous, but he sought to make clear assignments with well-defined learning objectives. For example, he provided a standard work set of instructions for each assignment, which takes the guesswork out of “What does the professor want here?” Emiliani says he’s focused on the learning objective and how he can help students succeed. “In a shop or office world, we don’t make it a mystery of how you go about doing the work. We have visual controls and we show examples of errors. So I provide students at the outset of the course examples of what the errors are and how things can negatively impact the grade. I set them up to get a whole lot of learning, succeed, and achieve one of their value propositions for the course in their education, which is to get good grades.”
· Load leveling: Rather than conducting a midterm or final exam, which is essentially batching information, students received weekly assignments. “Students really respond to that, particularly these part-time working professionals. They don’t have these big chunks of time to do these enormous assignments. They would rather have it spread out over time.”
· Visual controls/course remembrance: Students were instructed to create a visual control with pictures and words that synthesizes the essential things to remember from his course, and they were encouraged to use this at work and home. He also provided them with his visual control for what he hopes they remember. Emiliani says students appreciate the tools because they’re not sifting through books, texts, and documents to find something critical to their learning.
· kaizen: Emiliani led a kaizen review at Rensselaer to improve courses in the executive graduate degree program, which he believes is the first effort of its kind. “What we found is that the kaizen process—just about the same kaizen process that’s applied on the factory floor or the office—can be applied to improving courses with little modification. The summary is that kaizen is an effective process for improving courses, just like it is an effective process for improving set-ups, improving the flow of information in the office.” He says faculty enjoyed the kaizen experience and “thought it was a really good use of their time and yielded good results.” And, what’s more, kaizen is “fully congruent with an accreditation body’s desire for improvement in courses and programs. It is a specific process that puts meaning to continuous improvement instead of using the words in an open-ended fashion.”
www.lean.org Lean Enterprise Institute, Inc., One Cambridge Center, Cambridge, MA 02142 617.871.2900
George Taninecz is a business writer and editor, specializing in management and manufacturing issues, and vice president for the Manufacturing Performance Institute, a Cleveland, Ohio-based research firm. He can be reached at email@example.com.
Bob Emiliani is an associate professor in the School of Technology, Central Connecticut State University. He also writes about his lean work in higher education in “Improving Business School Courses by Applying Lean Principles and Practices” and “Using kaizen to Improve Graduate Business School Degree Programs,” which appear in the Journal of Quality Assurance in Education. He can be reached at (860) 558-7367 or at firstname.lastname@example.org.