Potential users of e-business need to consider carefully what they’re doing: simply adding the web isn’t necessarily going to solve anything.” So says Ian Farquhar, business improvement and IT director at Rolls-Royce combustion systems division, based at the Hucknall, Nottingham site. They need to look at their detailed business, manufacturing and supply chain situations first, he advises, in the light of what new, better IT-enabled processes can do. And his point applies well beyond ‘doing e-business’. He says companies may well be finding themselves torn between too many apparently separate, even conflicting, projects – like simultaneously trying to pursue lean manufacturing requirements, while still implementing corporate enterprise (ERP) upgrades and wrestling with e-business initiatives. How do you find a path through all this that makes sense for the business? Farquhar’s experience here is particularly useful: in a company the size and diversity of Rolls-Royce, there are bound to be strategic, company-wide proposals fighting for attention with individual business unit developments – and never more so than now, with the imperative to instigate e-business. And there are. Currently, Rolls-Royce at a corporate level is embarked on no less than four e-business projects across several of its businesses (including civil and defence aerospace and the associated energy companies), as well as an infrastructure programme involving data warehouse development, and continuing to roll out its SAP ERP implementations. Meanwhile, Combustion Systems itself has been focusing on reducing its IT systems to a single global engineering/technical system, shopfloor data management (SFDM) and latterly SAP, with advanced planning and scheduling extensions. It’s not easy, says Farquhar. “Improving operational efficiency is very important – eliminating waste, streamlining processes and getting lean manufacturing. But they all tend to be bottom-up driven and tactical, whereas putting in an ERP system and corporate e-business projects tends to be top-down. They have to be done simultaneously – we have to react quickly and do something with the e-business opportunity – so there is inevitably room for conflict. So that’s been my role: providing the link between tactical local implementations and corporate systems.” OK, so how do you do it? “You have to deal with it and remember that if [the corporate projects] are going to be effective they need to be appropriate to the coal face: meeting the needs of the business and the people that are going to be using them.” The bottom line, he says, is getting the corporate implementations – e-business or not – to complement the rest. Farquhar says that means you need to be very rigorous about understanding the current and future needs of the business – and what the intended systems will allow you to do, and what they won’t. “And the best way to do that is to involve the potential users – let them throw mud at it,” he quips. But there’s much more to it than that. “It’s critical to focus on ‘customer value’ and the ‘value streams’,” he says. ‘Customer value’ he derives from lean manufacturing principles as defining processes in terms of “what customers are prepared to pay for, or what the business needs to run.” The ‘value stream’, for him, is the “sequence of operations that you go through to add value in manufacturing and assembling a product from raw material to providing it to the customer.” Basic stuff perhaps, but these are the keys to bringing tactical and corporate endeavours, whatever they are, together. Looking at processes from these perspectives quickly reveals waste – and the areas where the metaphorical surgeon’s knife can best be wielded. At a macro level it means we’re instantly into internal and external supply chains: and Farquhar insists it’s the discontinuities here that most merit our attention today. He cites those between second and first tier suppliers, and internally between procurement, logistics and the various engineering and manufacturing departments themselves. The point is, he says, that with traditional supply chains, there are three immediate problems. “First, there’s a functional deployment of resources and systems around the internal divisions – so there are multiple measures and systems. Second, there are disjointed processes designed around each department for its own functional efficiency. And third, there are boundaries that themselves cause time delays and inventory build-up.” He cites one example: “We looked at the value stream for one component and found that for 65% of its lead time it was sat in a static inventory condition in the steps between us and our supplier, or inside our business. Only when you look at the whole chain, not just production, for example, can you see this,” he says. But he adds: “the technology is now there to do this much more easily.” And that’s a good point: with the relative simplicity and pervasiveness of web technologies and the power of data warehousing and integration, we can now make a difference. And we should, because the problems we’re living with that arise from this are manifold. First, notes Farquhar, efficiency is generally not measured as whole system effectiveness, and certainly not as whether the entire entity is building what customers want. Second, and worse, “the boundaries and different measures act as inhibitors to the flow both of business processes and manufacturing processes – and there are duplicated activities.” And he gives the example of building geometric models in engineering for stress testing – and then recreating them in manufacturing systems for process planning, operations and so on. “Why do we do that? When you take a customer value or value stream perspective, you realise you should be using the same tools and doing the data entry just once. It not only cuts out the obvious delays but, just as important, the errors you introduce.” It’s clear that Rolls has got the measure of this, and certainly Combustion Systems is focusing hard on moving away from it as fast as possible – and also harnessing the initiative specifically to leave behind the traditional ‘push’ style of manufacturing and supply chain management in favour of ‘demand pull’. Farquhar describes the old ways thus: “It starts with a push to make ASAP; then it trundles through the system in a semi-planned manner, and then you complete when it gets urgent with the usual expediting … And you find that means you’re adding value 5% of the time; 95% is just cost with no value. It’s ‘work in business processes’ if you like.” Again, beating this is about “creating a lean value stream,” says Farquhar. And how do you do that? “Get rid of the boundaries,” he says. “It’s a matter of challenging the traditional functional roles: then you can look at changing to ‘customer pull’ as the trigger to start work and run with it all the way till you deliver on time.” The result: “you cut lead times and inventory.” And a whole lot more. This is what matters, he insists – streamlining the supply chain. And to do it, Farquhar repeats, “first you specify customer value, then identify the value stream for every major product and business process – and simplify, map and understand it so you know the workplace reality in business and manufacturing terms.” Otherwise, he says, real lead times for products remain unknown, machine operating times versus wait states are unknown. “Remember, coal face reality is often very different to what the business thinks is happening,” he adds. “Then you can eliminate wasteful activities, like wait states and transport,” he says. “Ask the question – does it add value? If the answer is no, get rid of it.” With that done you can create the flow and align your resources and systems to the new value stream. “Applying any IT then is a matter of viewing it as an automatic support tool. You deploy it in a ‘process-based’ way to maximise effectiveness,” he says. Farquhar is adamant: not only is it is possible to go all the way with this – it’s worth it in spades. “If you focus on the value stream and reducing lead times for all the activities it is possible to move the whole operation to a ‘pull’ mechanism. Then you can lock in close with MRP only for long range capacity planning, while synchronising production on pull-through, and cutting lead times to just the value-add part.” Plugging into benefits And then the benefits flow. “Remember, the degree to which you need to forecast reduces if you do this, so you don’t need MRP at the shop floor level any more,” he claims. And he cites Mars Corporation as an example where MRP is only used for long range forecasting and “they’ve switched off all the SFDM (shopfloor data management) and exchanged it for physical and visual controls.” Further, as you reduce lead times the scope for additional simplification of processes also increases. For example, with ‘pull’, “you fix the lead times so the level of inventory is reduced and that means that tracking is easier too,” says Farquhar. And he claims that going the lean route in this way has made a significant difference for Rolls-Royce. In one case, he says, the cycle for complex fabricated combustion liners was reduced from 44 working days to 19. Ultimately, the goal, he says, is to have a means of controlling workflow through the factory that’s fast and completely foolproof. “Don’t get me wrong,” he says, “we’re very much applying SFDM for manufacturing planning, but historically we would track every single operation, even if it was only two minutes – and the transactions could take longer than the processes.” He gives the example of a laser cutting machine. “Over 1,000 part numbers are cut on this machine and it gets through them in seconds. It was taking the SFDM 20 minutes to launch product into inventory, so the IT was a big bottleneck. Now we have a more gated system that tracks movements after several operations.” Meanwhile, on forecasting, there are similar gains. “One of the biggest challenges is that suppliers won’t allow you to drop your orders straight into their MRP systems because there’s too much variability, too much noise,” observes Farquhar. But with a streamlined, ‘pull’ mode supply chain, as lead times reduce and inventory is slashed, contrary to received wisdom, variability actually decreases. Farquhar explains: “As your dependence on forecasting is reduced so is the need for complex forecasting logic buried in systems to get around the earlier problems of very long lead times from the ‘push’ systems. It all gets simpler.” In summary, it’s clear that to get real benefits from supply chain IT you need to rationalise corporate and tactical IT teams – and ensure that process designers and system specifiers are in touch with the real – and the desired – manufacturing and business situation. Focusing on customer value, mapping the current value streams and challenging and simplifying current processes and assumptions are then critically important. This done, you can align resources and systems to the value streams – making sure that the IT itself and its functionality are transparent, easily updated and flexible enough to respond to change as it happens. Says Farquhar, “As the advice from Masaaki Imai, a Kaizen guru, goes, it’s not ‘what we know’, ‘what we had time for’, or ‘what we liked’. It should be ‘what we needed’, ‘when we needed it’, and ‘we learned new skills when we needed to’. Good luck.”