Big picture lean thinking ain't always that easy

Taking lean thinking beyond the shopfloor can be a considerable challenge for any manufacturing organisation. It's not that the methods are difficult, nor that the IT requirement is especially onerous nor expensive: the issues have more to do with people, company culture, change and perceived overall cost and effort, versus jaundiced views about prospective business benefits. People – managers and operators alike – have seen it all before. It's well worth looking at one well known UK automation equipment manufacturer, that wanted to remain nameless, to get a good feel for some of the problems, some of its solutions and some seriously useful pointers to making a success of what should be an essential and all encompassing initiative for most of us. The company in question has already developed and implemented an impressive spares ordering website which demonstrates well the power of IT to assist with lean execution well away from the factory. "We defined a set of rules by which spares can be purchased," says its engineering manager, "and the website takes orders straight onto our ERP system, onto the shopfloor, which in turn drives our supply chain. We know our costs in terms of parts and production operations; it includes a delivery guide; and it's flexible." What's more, the spares site provides a lot of component and commissioning information, which allows users to choose what they need accurately, and self-serve installation documentation. All of which sounds excellent – and our man now wants to apply similar logic to the company's main production systems to get them running equally smoothly and also, specifically, demand-driven, thus reducing costs and lead times and getting away from the traditional fire-fighting. It's a good idea and his thoughts are founded on robust lean principles. How to do it? He wants to implement, for example, what he describes as collaborative tools in the form of advanced planning and scheduling (APS), possibly Preactor software, as well as portal systems that can publish detailed demand requirements to suppliers in real time, and provide for automated call-offs and supply chain event management (SCEM). "We're looking at APS to help provide a new infrastructure and a new interface for our users that will work on top of our existing ERP," he says. "If we apply that to the supply chain, then the suppliers would see daily delivery schedules so they could plan to supply better. Then, as the information gets used internally, and becomes more accurate – which comes back to the BoMs [bills of materials] and MRP data all being accurate and used properly – everything gets better." Compare that with today. "The difficulty, at the moment, is having an infrastructure that works in batch mode," says our correspondent. "The problems around getting parts in on time to match production schedules, for example, are due to batch-run MRP assuming infinite capacity, allowing back scheduling and the rest. So if a late order comes in, it'll tell you that you should have bought the parts three weeks ago. That confuses our suppliers and, because we don't change promise dates – to avoid turmoil in MRP – it in turn puts pressure on our purchasing department to deliver, say, 10 week lead time items in five weeks." Which is a problem. "Buyers become expediters," he says. "If we could just use MRP for the long term planning and replace it with something that gives them the right detailed information first time, we could solve that problem… Kanbans could take some of the parts out of MRP control but that really only works for the runners – and we have very few of those." It's a common story throughout British manufacturing. And so is the fact that although management may recognise their systems' and processes' shortcomings, getting a business to change is easier said than done. And it's not just a case of dogged resistance to change, nor necessarily of laggard mentalities. At the production end, for example, our manufacturer is already operating lean methodologies. The issue is almost always threefold: first, not everything is wrong; second, there's more than one way to skin the proverbial cat; and third, company culture gets in the way. "The existing systems and processes are completely embedded in people's mindset," agrees our man. And he adds: "Our underlying ERP is reasonable: it's the way we use it." But a lot of effort and investment went into building it and the associated processes: it works after a fashion and it's a case of 'better the devil you know'. "We've been at a crossroads for some time," says our man. "Do we change the ERP, which would cost a lot of money and upheaval, or do some business process mapping and re-engineer to the processes we identify we need – and then undo some of the workarounds on the existing system? We've had a bit of a problem getting to grips with it." And that's despite the fact that the company's subsidiary in the US has gone the latter route, changing to strip out non value-adding processes, and is now starting to see significant improvements. Looking in brief detail at a couple of points is particularly helpful here. The company has a sophisticated product configurator built into its ERP to handle the massive potential for variety in its products' size and specification, and our man says it's 99.9% accurate. But the problem is less the configurator and more when it's run – because at quotation stage it isn't. Sales uses a separate quotations system to quote a high level price and delivery date. Only when the order comes in does a contracts engineer generate the detailed spec in the configurator that then creates the BoM – and the production system and MRP don't see it till then. Conflicting priorities Hence some of the problems around parts and lead times – and it's an excellent illustration of the power of modern IT for change in lean terms. When this dual stage system was devised, it was the best available. The company has since looked at more modern configurators – effectively experts sitting on sales people's shoulders, speeding right first time order entry and validating high variability, multi-option products, while simultaneously saving engineering and production time. But there's a price to pay to get over the current duplicated processes and problems down the line. Then when it comes to order promising, and production and supply chain scheduling, again there are limits due to the age of the system. "We have very basic capacity planning, but our ERP doesn't recognise product types and orders, and it's just planning in weekly buckets," says our man. So there's no finite scheduling and no ATP (available to promise). "We'd like to be able to take an order and say when we can make it, with the system picking up parts with long lead times and examining if we have stock cover or if there are outstanding purchase orders, in real time." The point: integration and the timely availability of information (so also workflow) are as important in lean thinking as they are in any business thinking where multiple roles and interests are concerned. But while some in the company are bound to clamour for ATP first, the supply chain remains a key issue – and it's easy to see how companies can let time slip by as they debate the inevitable options and weigh the cost/benefits and priorities. Business intelligence, for example, can make a big difference; simulation and 'what ifs' likewise; and alerts and exception reports can be key in bigger picture lean. If the market changes suddenly, for example, it's going to take time for the triggers to get back through production cells: far better if you can signal instantaneously. For this company, however, it seems likely to tread the APS and web systems path. "Longer term we'll look at using different reporting tools – look at the assemblies for build this week and see the shortages, for example, and extend that to the suppliers," says our man. For the rest of us, lean value stream mapping and a good dose of logical thinking are the starting points.