Spreading lean thinking throughout British manufacturing is a survival issue. Brian Tinham reports from the latest MCS Forum
Lean is the obsessive elimination of waste," insists Paul Mason, managing director of lean consultancy Hosca, defining 'waste' as what customers would perceive as processes and actions that don't add 'value'. But it's not just that: "It's getting more from less," says Andrew Rumney, lean solutions director at Microsoft Business Solutions VAR eBECS. And he goes on to describe the basic principles of 'flow' and 'pull'.
We can all understand the intent, but what then makes lean thinking more than simply the application of common manufacturing and commercial sense? What makes it the sustainable success that it seems to be for companies – some very high profile – that practice it, and claim lean as the centrepiece of their ongoing improvement strategy? How does it relate to other initiatives? What are the secrets here?
We put 30 senior manufacturing users from across industry, some consultants and a dozen IT vendors and service providers round a table and challenged them to come up with answers to help UK manufacturing plc learn, and get the benefits clearly there to be had from 'going lean' – and not only in terms of cutting costs, but becoming more responsive to customers. This is a summary of their frank and open dialogue, following on publication of our in-depth 'Lean Thinking Boardroom Report' on the subject, which covered everything from improving business processes and manufacturing and supply chain practice, to implementing appropriate IT at every level to make corporate 'lean' work.
First, what is lean, beyond the initial, vague definition? Well, with its dual focus on waste removal and customer satisfaction, it's a strategy for survival: more than that, it's an initiative to promote manufacturing business health, providing top-down and bottom-up routes for achieving the desired goals, and prioritising, managing and executing continuous improvement.
Crucially, it includes a proven set of methodologies, tools and, increasingly, preferred IT that together amount to a formidable armoury able to take care of almost any manufacturing requirement at any level and any stage. But, "it's not just a bag of tools," as Will Seager, divisional director of RWD Technologies puts it. "It's the methodology and brain power behind using those tools – how we manage the elimination of waste… The worst thing to do is call it 'the Toyota Production System', 'the Ford Production System', 'the Jaguar Production System'. It is an operating mentality."
It's important also to note that it's not just a fancy title for common sense – indeed, those that do lean properly are successful because of its holistic approach and its structure, and there are those who say, 'why would I bother to investigate other initiatives knowing this one works and has got sticking power'.
It's also pragmatic – while it prefers zero inventory, single-piece flow, rapid changeovers, zero duplication and so on, lean practitioners recognise that production necessarily involves some batch processes, that commercial pressures (from raw material availability to customer service) influence purchasing and stocking strategies. The direction, emphasis on systemic improvement and tools and methods are the keys.
How far can it extend? Ralph Woodhead, lean consultant with Open Logistics, suggests that, particularly in the automotive sector, it's about whole supply chains, virtually from mining the ore to delivering finished items. He's not saying that's practical, but it's the end game. And Chris McKellen, consultant with Manufacturing Awareness, advises us to at least think: "from the customer to the supplier's supplier, and everybody and everything involved in the business – certainly not just lean manufacturing."
A couple of caveats: given the scope, you need ultimately to get the entire company working with lean – and that means culture, education and, absolutely key, total management commitment. As David Tudor of Oracle says: "It's sustainable as long as you have the processes and management in place, otherwise as soon as something else more exciting comes along, management's eye will be taken off the ball, and in a year's time they'll be saying 'what was that lean programme?'." In fact, the best lean manufacturers also have a roaming lean improvement department, always able to provide direction and support.
Also, as Clyde Bennett, solutions manager with Microsoft Business Solutions, observes: "The bottom-up approach is typically a good way to get started through the departments. And 'value stream mapping' [lean's key top-down start analysis] is the modern equivalent of business process re-engineering done by a small top team looking at the whole value stream. The benefit of the latter is it avoids islands of excellence – or fixing the wrong problem in the wrong order."
Best advice: by all means do something, but don't dive into one part of the value stream or one department unless you're sure about the big picture. To quote Bennett again: "If you're working from the bottom-up, which a lot of shops are because they're not getting support from top management, then you can focus on waste in sub optimal ways, which is somewhat risky, whereas if you go from the top-down, you have the opportunity to look at the bigger processes."
What success can you expect? Mike Collins managing director of coolers manufacturer Hydropac, and also lean coach with Cranfield, says: "One company I worked with in the States, since we introduced lean in September last year has increased performance in raw materials and taken $7 million out of inventory – that's 27% and that's in 10 months… It's very sustainable: they've got lead times down to two days on raw materials … and the warehouse is very empty at the moment. In fact, they've managed to get rid of one of the offsite warehouses and all the associated costs."
And he adds: "Yes, you want to have support from the top of the business, but you don't even have to do a full value stream map. You can identify you've got a problem with raw materials: most companies I work with find the reason they can't get product to their customer isn't because their people aren't working hard or because the workshop's untidy – it's because they've run out of materials."
What about IT?
So what about IT? Nick Lewis, project leader at the Basingstoke bit of ITT Cannon, was faced with the objective of getting to a one-day cycle time for order processing and production, with immediate customer delivery date confirmation, more efficient use of the assembly cells and the potential for same-day make-to-order. He makes the point that what seems at first sight impossible, may well not be – particularly where most components are kanbaned and even where they fall under MRP.
His solution was a finite capacity-based local cell scheduling system on top of the firm's existing AS/400 ERP, with screens for sales and screens for the cells updated automatically. It uses hyperlinks into the AS/400 for order details and the rest, runs semi-independently of the firm's daily MRP, and provides visibility for sales and the cells, as well as direct translation of sales orders into works orders. Cost of the additional software, he says, was low; it was up and running quickly; it didn't involve modifying ERP; and, he says advisedly, it was tested and rolled out cell-by-cell. It's an aside, but he adds that the system is now being augmented to provide similar functionality for the firm's occasional non-kanbaned components, managed under MRP.
Could he have done it without IT? "I think we'd have struggled. The great thing about it is it's so up-to-date." But it's not just IT. "Line leaders now are people who decide how the production process should be working, and we're also getting feedback from them on things that don't make sense. So it's helped us make a connection with the shopfloor and taken a level of bureaucracy out."
It's been an immense success: immediate improvements in productivity, customer service, inventory position and so on – to the extent that it's being taken up at other ITT sites globally. Lewis does, however, concede that some of the IT could have been done differently.
MRP's different roles
Collins, for example, asks why kanban-controlled components were removed from MRP notice: "If MRP does its job properly and kanban's doing its job, the parts will be there and MRP will see them and won't tell you to make some more." He also makes the point that with MRP still in the picture, it's a protection mechanism for lost kanban cards and the like that can screw up what's otherwise a simple system. All of which Lewis accepts, although he indicates that ITT wanted to prevent the slide back to 'comfortable' MRP from kanbans – hence the removal so there was no way back – and that barcode kanban signalling, with auto emailing to buyers or PO generation for suppliers, was introduced to minimise problems with lost or forgotten cards.
Other observations: Seager reminds us that capacity analysis-based systems also help reveal anomalies: "You'll often find standard times on the system bear no resemblance to the reality including changeover." IT system integrity is where companies often suffer: how good is the data and how well is it being updated? And David Buttery, manufacturing and supply chain systems manager at Ford, suggests that processes, people and training, particularly at cell leader level, are also key here. To which Lewis adds, involving IT as early as possible, "not deciding what you want, and then, when you've got the solution, going to the IT department."
Incidentally, this kind of initiative could also be taken further – for example, linking sales in even more tightly and allowing movement of important customer orders around against real available capacity, with 'what ifs' and the rest. Transparency based on one set of live information is a powerful tool. The only concern: bespoke software and add-ons are all very well, but watch for release management, maintaining integration and look and feel.
Moving on, however, there are other opportunities – and other ways. Mike Lilly vice president of Lilly Software Associates: "We've had success in the States with a couple of dozen companies in implementing an approach that cuts work in process [WIP] in half, delivers lower lead times, approximately half, increases turnover 20 to 30%, takes on-time delivery from horrible to as close to 100% as they want to get — and these are results we've achieved on average in 30 to 60 days with no value stream mapping whatsoever."
Interested? "We use an approach called 'simplified market pull', which is protected pull. If you're making-to-order and delivering in 12 weeks, pushing work out on the floor, you have way more work out there than you need, and tons of fat you can cut. So if you want to deliver in six weeks you stop releasing work for six weeks: you don't release work any sooner than the lead time you want."
And he makes the point that slashed WIP means cash to fund improvement projects, while reduced lead times and guaranteed on-time delivery opens up valuable sales and customer service opportunities. As for how the shopfloor knows what to work on next, he says: "It's based on what percent of budgeted protection, in this case time, has been consumed thus far… So when a person finishes a job, they look at the queue and pick the job that has the highest percentage of its protection used up. It's an electronic protected pull system."
Sounds a lot like Theory of Constraints? Yes, and it is – but the takeaway is that going lean from a shopfloor management and IT perspective requires some rethinking – and certainly reconsideration of MRP and its role, not least because of its roots in to-stock manufacturing as opposed to lean's drive for to-order. Indeed Lilly says: "My dad, Dick Lilly was part of the team at IBM that wrote the Pics manual, the basis of all MRP systems, and he has publicly apologised for that!" Humour aside, Lilly junior makes the point: "When I go into our customers … I admit that it's 80% about changing the way you release and manage work to the floor and 20% about software."
And hidden in that statement is, whether you're looking at his simplified market pull, or straight lean, MRP can't be allowed to get in the way of factory scheduling or customer service – although it still has longer range planning and procurement and materials protection roles. As Lewis says: "For lean to work, it was a matter of breaking away from the MRP culture." Indeed: eBECS' Rumney is a little extreme, but he and most lean advocates encourage starting from the position that there isn't a place for MRP at all – and then working back to it for the lead time and market-constrained items, but implementing kanban (IT assisted or not) wherever possible for its simplicity and inherent forcing of 'flow'.
All of which rings bells – and throws up difficulties. For example, Colin Little, operations director at Elkington Gatic in Dover, which is expanding its business in steel fabrication, wants to move control from MRP to capacity-based planning with automated shopfloor data collection (SFDC) in the lean way – but says that with business booming and logistics issues, the day job gets in the way.
And John Bentley, joint managing director of LB Bentley, which makes high pressure filter equipment and sub-sea valves (and is currently tweaking its Lilly ERP system to improve efficiencies and customer service), comes back to the data issue. "You're only as good as the data you're putting into the system. It's a long, long process: we now know, for example, how long it takes to make things and that's very interesting compared to how long we estimated – and hence finite scheduling is badly impacted."
Moving from push- to pull-based manufacturing, although a key management focus, he says, is going to be a long haul. And it's not as though this is the first initiative: the company has already instituted substantial change with finite scheduling to get due dates and quotes under control. It's all fundamental stuff, but the point is that the real world gets in the way. Inevitably it's a management issue, but the bottom line is, to survive and compete you've got to do this. And it's worth pointing that out to the entire staff – understanding that, whereas with ERP implementations burning the midnight oil was for middle management and admin staff, the challenge now is that it's for the whole business, including the shopfloor.
Wrapping up on this, it's worth dwelling on that word, shopfloor. Mark Sutcliffe, managing director of MVI Technology, points out that ERP, MRP, even scheduling systems, are fine – but that on the shopfloor "you're dealing not in a land of transactions but of events – people making decisions based on their experience and their knowledge. And raising their knowledge, raising their awareness, allows them to make better decisions. SFDC done properly lets them record the reasons why machines are failing as well as the fact that they fail: why we've got waste or the reason why we've accumulated buffer stock and so on."
He tells of the Coca Cola Schweppes initiative of more than a decade ago – "WAP 2000: what's appropriate counts" – reminding us that there is no place for dogma in business. But he insists that if you rely on informal systems, or put it another way, inadequate systems or no systems at all on the shopfloor, you're not going to get the best decisions or efficiencies, no matter what the corporate initiative. You'll promise high and deliver slightly under.
"If I know that action I'm about to take, or not take, will cause the order to arrive at the next position late, and that directly affects a key customer this week and we'll incur a penalty, I have the opportunity to respond to that," says Sutcliffe. Lumping together shopfloor systems as 'SFDC' is a mistake: there are simple systems and sophisticated systems that are fit for purpose. MVI's, and others like it, are automated but, crucially, also encourage and force operators to enter reasons for exceptions, link that into analysis, and provide two-way communications so that the shopfloor is in touch with the business – and middle management can make a real difference too.
Final thoughts and advice: Forum agrees that lean thinking should be used well beyond the factory floor, and not just out into the supply chain either – although these are key places to cut your teeth. Beyond that, Angie Coombes, lean engineering executive at Ultra Electronics Controls, demonstrates how lean thinking, supported by modern IT, is a powerful approach everywhere – from the MD's office, through HR, into purchasing… Read her thoughts and achievements on the website. Above all though: people, training, culture, systems, communication, management commitment and appropriate performance measures all need to be top of mind.