Vital statistics

In September, the entire workforce of Thorn Lighting received a gift of wine and joined in a company-wide celebration. It was a heartfelt thank you from its management for the supreme performance that crowned it Factory of the Year at the Best Factory Awards (see WM October). The Spennymoor plant, part of the Austrian Zumtobel Group, won because it sets dazzling standards in quality, cost, customer service and relentless continuous improvement. And it counts the combination of lean and Six Sigma as the most powerful accelerant on its journey to excellence. So let's look at why these methodologies are so important, firstly from the viewpoint of a global consulting group and secondly from Thorn Lighting itself. And let's make life easier by choosing 'lean sigma' from the confusing mass of terms currently in use. First of all, Richard Holland, UK MD of TBM Consulting Group – which is messianic about its efficacy – puts the basics in place: "Lean and Six Sigma look at fundamentally different things. Lean looks at the waste in the process – the steps in the process and how to shorten them, and reduce lead times. Six Sigma looks at variability – the measurable output." He compares it to a journey to work; lean asks how he can get there by a simpler, more direct route; sigma looks at why it is possible to arrive early one day and late the next. It starts people studying what they need to do to get a more repeatable process. "We preach that you really do need to start lean first to simplify the process because that is often the solution in itself to the problem of variability. There will still be some processes or business issues that need the heavy analytical approach of Six Sigma but you will have thinned them out." TBM usually finds that Six Sigma's underpinning methodology of DMAIC – define, measure, analyse, improve, control – is needed for processes that are invisible to the eye: "You often need tools like DMAIC and regression analysis to solve, for example, something happening inside the tanks in a brewing process. Lean can't help you there." In a one-year period with a client, Holland says that TBM would typically run only one or two sigma activities among 30 lean activities. "Rolling out Six Sigma the GE way [one of the earliest and strongest adherents] with multiple black belts is not a methodology for engaging everyone." He feels that a more collegiate approach works better for routine improvement: daily management through KPIs and team meetings to discuss improvements, and the use of kaizen to make them happen. "That involves everyone. Then there isn't a problem with having a specialist in the business who can bring all the complex tools and techniques to bear in solving a really difficult question. It's a bit like bringing in an external consultant!" Now let's hear from the man who turned the theory of lean sigma into a living reality across the Spennymoor plant, operations director Terry Carmichael. Spennymoor embarked on a radical change programme four or five years ago. At corporate level, there was already firm involvement with lean but not with Six Sigma. With the arrival of Hans-Jürgen Thews as chief operations officer in 2006, that changed to a company-wide adoption. By then, however, it was already underway at Spennymoor. Carmichael had seen it in action in his previous working life with Philips, auditing plants around the world against the Ford Q1 standard. He was convinced that the marriage of three methodologies – Hoshin planning, lean manufacturing and Six Sigma – produced far greater results than sole use of its constituent parts. He is adamant, however, that the labels aren't important – it's improvement that counts. And gaining people's interest is fundamental to that improvement. Carmichael uses the cost-of-failure model to determine the real priorities – something he calls 'the vital few'. "I would recommend this to everyone. The process most people use is to look at last year's performance and set an incremental improvement without understanding that it might not be a great target. We go right back to basics by asking what cost would we be able to drive out of the business if we were excellent at everything?" By looking hard at the potential of, for instance, 100% delivery versus 100% labour performance, he is able to identify quickly the things worth working on in the next year to give the biggest return. "When most people have limited resources, it's so important to understand that you are working on the vital few rather than the trivial many," he explains. The figures are indisputable. Take just one example. In 2004, Spennymoor's labour performance was running at 55%. The gap between 55% and 100% was a cost of £1.1m. Today it runs at 92%. So how did Carmichael and his team tackle the vital few? He is in no doubt that the primary lean tool is kaizen breakthrough (KB). Every month, line by line, Spennymoor hit the shopfloor with a KB programme. Typical results were a 20% increase in productivity, a 40% reduction in floor space and an 85% reduction in end-of-line rejects. Note he used kaizen, not Six Sigma at this juncture. It could almost be Holland speaking: "Lean is about setting up the processes in the right order, taking the waste out and delivering products to customers when they need them at the right quality. Six Sigma is about looking at individual processes in that supply chain and getting the best yield and first time pass through. The first is about putting the processes in order; the second is about making sure you are getting the best out of every process and every step along the way." In a typical KB event, Spennymoor stops the target line for a week: "We plan it so we can afford to take it down." Most of the operators are deployed elsewhere but two or three are retained to work as part of a KB team of about 10. The rest come from finance, marketing, maintenance, etc – a deliberate way of gaining an end-to-end view. A lean sigma black belt facilitates. "We are looking to bring in people with fresh eyes who have knowledge of the business along with the people who run the line themselves," explains Carmichael. They are all trained for the first four hours in the tools and techniques of KB, using a package tailored to the week's projected activity. He says that simple tools like pareto analysis and process flow are sufficient for around 80% of the time. Complex sigma tools like design of experiments come into play for the intractable 20%, led by the experts. In many cases, the join between the two methodologies is pretty hard to spot. The same team may start by taking a kaizen approach to a manufacturing process but then use sigma workshops and techniques to solve problems thrown up along the way. For instance, a problem with one of the mould machines was producing a 10% reject rate on plastic parts. The more specialist team members designed experiments to work out the best settings. Others embedded the new parameters into standard operating procedures, machine set-up documentation, engineering drawings, quality processes and maintenance routines. The problem has never recurred. As an aside, Carmichael points out that companies often neglect to integrate sigma improvements into quality systems which "are very much about say what you do, write it down and prove it with documentation to auditors. If you start to introduce sigma methodologies that are at variance or not recorded, you could be in trouble." Similar approaches may be seen in other plants. Where Spennymoor differs from most, however, is in the degree to which lean sigma is integrated into everyday activities rather than – as in so many companies – making sporadic guest appearances. It stems from a radical reassignment of responsibilities. Three years ago, Carmichael had a head of industrial engineering, two people working on CI, an engineering department, a few black belts plus quality engineers and inspectors all working as a company wide resource. Today, he has a single department of just eight people to cover the lot, all of whom are qualified as lean sigma blackbelts. With some government support, each has gone through extensive training, accumulating a full suite of skills in lean sigma techniques, industrial engineering, and quality control. "We invested heavily and they are very, very skilled people," he explains. Instead of working plant-wide, each of them has complete responsibility for two lines and everything relating to them including suppliers, production and customer delivery, including warranty returns. It goes further. Spennymoor is working with Sunderland College to take all operators to yellow belt standard through a training programme delivered and assessed on the shopfloor. Others like production planners are working towards their green belts and the goal is to have 100% of employees educated to some level of Six Sigma. "A lot of people train in it but they never get the simplicity of what it's trying to achieve," reflects Carmichael. "Firstly, it gets a structure into people's minds that lets them tackle any issue – we issue a little card to everyone which details DMAIC and the 12 steps within it. It also brings a common language. So when we talk about critical to quality [CTQ] dimensions of a product or service, people immediately understand what we are talking about. We have made a route by which everybody can contribute at some level." It costs 20 hours per operator to go through training. Carmichael feels it plays a vital part in engendering a culture of acceptance: "It's rare that we meet strong resistance to improvement. We are pushing against an open door." He believes the start point for anyone embarking on lean sigma is understanding the effect of change on their own plant culture: "I've been in organisations where cavemen sit – you have to spot the Citizen Against Virtually Everything. Managers must transmit their desire to the people who are going to carry out the work." He has some brisk rules of thumb for success. We've discussed prioritising through cost of failure. He and Holland are united in believing that when companies choose lean sigma projects that are not important to the business, no-one reviews them and they just die a natural death. But even with the right choice, deployment isn't easy. He is a firm adherent of Hoshin planning, a methodology for deploying overall strategy into a set of clear actions and objectives, cascaded to each level and function within the factory. He is clear about the importance of picking the right improvement tool to avoid demotivating people: "And in the midst of all this wonderful complexity, if you can use the most basic tool, do so." He knows he is at variance with accepted thought in making the lean sigma black belts responsible for the results: "Many choose to put them in as consultants rather than someone who is part of the business. But when you do that, the rest of the organisation comes to resent them. Anybody can identify what's wrong – putting it right is the difficult thing. And I have seen people put into a position where they just become the critic rather than the help." He is clearly an advocate of the team approach but, regardless of the chosen route, he maintains it is always important to have a clear overall message of where you are trying to get to. It could be a very simple one; for instance "I want an uninterrupted supply chain." Or it could be the more complex one he recalls from his Philips days: "It was just 'I want a 30-euro TV tube.' But it was really effective because it was something simple that people could understand. It wasn't highbrow or 30 pages long." On the wall of his office is an ancient saying. It summarises something he passionately believes in: "If you want the people to build a ship, do not ask them to collect the wood and nails. Simply create in them the desire to sail the seas; the rest will happen by itself."