Look before you leap

When times are tough, small steps for survival may not be enough. Better inventory management may keep you afloat a bit longer, but a radical change to order-driven manufacturing or some new plant could be much better. But how can you be sure? Industry is awash with companies that jumped, and landed on quicksand. Ask the huge confectionery company that missed Christmas because it underestimated the effect of its new ERP system on its supply chain. Ask the auto supplier that introduced JIT (just in time) without accurate demand planning: its Sri Lankan supplier could turn on a pinhead but it couldn’t do much to close the shipping miles. Have you heard of the chemical company whose Asian suppliers met reduced lead times by missing out a vital stage in QA? Two weeks’ production rejected. And what about the vehicle supplier which slowed down its entire production – and missed due dates for six months because its interior fitting line couldn’t keep pace with its re-equipped body shop? Clearly, a little forethought pays dividends, but since it’s not always easy to assess the impact of change on every area, computer simulation seems an obvious route to getting the big picture without the pain. It works for Ford: the company analyses proposed changes to in-line vehicle sequencing on a two-yearly basis using Powersim simulation software. That means it can avoid renovating a factory and disrupting production until it’s sure it will get tangible benefits. In one study of materials flow, the simulation demonstrated that the proposed sequence changes would speed in-line delivery, raise process stability and reduce work in progress (WIP) inventory. But it also highlighted increased opportunities for errors in material delivery in more complex vehicles, and showed the ripple effect this could have. So the technique was used to test multiple options to avoid the problem of over-simplification. When that turned into a real-life implementation, it resulted in $800,000 savings annually in one plant alone. Company-wide, it is expected to save Ford more than $200 million in production costs over 10 years. Ford’s Vic Leo says simulation helps managers across disciplines focus on what adds most value by reducing a complex subject to its essence without dropping the detail. But will it work for everyone? Over years of reassessment and modification, companies like Ford have a clear handle on what pays and what doesn’t in an established ‘demand pull’ environment. It has the data to feed sophisticated techniques like this. Few others are in this enviable position, and certainly not those SMEs contemplating radical changes. They can probably use techniques like this later, but not as a first step. Software not the way? Decent software vendors agree. Dr Maz Mohamedali, of ERP supplier Lilly Software, for example, is a firm advocate of the simple approach when rethinking information flows. Take changing from paper-based invoicing to electronic approval. “It goes to accounts payable for date stamping, then to the department who ordered it, then back to accounts payable for the next payment run and the supplier eventually gets the cheque,” he observes. And plainly, most can see the value of a simpler alternative where goods or services, once delivered and electronically matched to an order, get paid straight away. “Would you use our software to simulate that? No. Would you use anyone’s? No. Would you get people round a whiteboard? Yes.” Here’s another surprise. Lee Bamber, senior management consultant with ERP giant SAP, favours paper, pens and post-its for the early stages of any major change to business processes. “I have worked with so many senior managers who’ve said ‘I didn’t know we did it like that’.” The paper stage helps everyone to participate and it provides a shared and accurate picture. ‘Lean’ specialists have been saying for years that value stream mapping (VSM) should lead any improvement programme. It charts the dependency of every stage in internal, supply chain and factory floor processes and examines them for the contribution they make to the business. It is the first step to eliminating wasteful or redundant processes. Bamber uses the Supply Chain Council’s SCOR (Supply Chain Operations Reference) model, which is similar, but works at a higher level. He maintains, however, that there is room for both. Lego and automotive supplier Mann & Hummel have both gone the SCOR route for this. But after that, Bamber is adamant that computer-based support can and should play a part. He advocates a three-step approach. First, VSM/SCOR: find out where you are in terms of suppliers, lead times, the major steps for satisfying orders and so on. Then see what’s possible: why don’t we do it this way? This could well involve simulation to understand effects, costs and benefits. Third, implement: since this may involve new systems and technologies, a third kind of business mapping is needed, to allow you to explore what precisely will best take you where you want to go. Stage Two tends to involve specialists like Lanner or Powersim, with which SAP for one has an alliance. But simulation is still regarded as a black art by many, so who do you involve at the nitty gritty? One major confectionery company employs a floating expert who follows process flow, partly to capture what happens but also to gain employees’ acceptance. His biggest issue is convincing users that real benefit can derive from such an arcane process. For this reason, some simulation firms mix modelling services with software. Simulation can, however, work very well. For example, it contributed to a 15% reduction in unit costs at the 250-strong Newmarket plant of photographic processor ColourCare. Lanner’s Witness software tools were used to analyse plant operations and to predict the interaction between processing volumes and staffing. With huge fluctuations in demand for overnight processing, staff rather than capacity is the key. Even a small miscalculation can raise processing costs and impact tight margins. Although Lanner created the core model, ColourCare’s own production managers enter the data that drives it into an Excel spreadsheet interfaced to Witness. Details of demand and despatch times are fed in daily, and it’s possible to see not only how many staff are required, but also at what stage of processing and time of day. The model provides analysis of the cost per hour and cost per order for any parameters. The technique has also influenced longer term development. By simulating staff activities, it became clear that greater flexibility in allocating tasks during a shift would help overcome bottlenecks. Staff were therefore retrained to cover each other. Simple. It’s worth pointing out, however, that specialist software is not the only route to predicting the impact of change. Companies that have advanced planning and scheduling (APS) can use the systems to resolve longer term organisational issues as well. Mettis Aerospace’s Redditch-based Machining Division used this technique to justify major changes to its shop floor. It invested £2.5 million in new equipment and facilities, including high-speed FMS machining centres, but before doing so, used Lilly’s Visual Manufacturing ERP system to cost-justify the outlay against business from Rolls Royce for the Trent 500 aero engine. Scheduler controller Ray McCoy says that accurately simulating and understanding the impact of the new business was surprisingly simple. First, staff loaded and ran projected sales over a five-year period against existing production capacity. “It showed some massive peaks four years into the project which we simply would not have been able to accommodate,” recalls McCoy. By running ‘what-ifs’ against new plant options, Mettis was able to arrive at the optimum combination. APS played an important part in proving to Mettis and Rolls Royce that it was geared up to take on the work. “I’m a big fan of this system,” says McCoy. Feeling threatened? Bamber can cite several instances of SAP’s APS being used for similar long-term planning. He’s recently worked with a food company that, having run the long term order book against available capacity, realised it should cancel expenditure on one piece of plant and apply it to another. “Because we have capacity planning within the software, a form of simulation can be run whenever the business appears to be undergoing changes – slight or significant. It lets you increase your forecast, say, by 20% and assess the impact.” What happens, however, if the change is so radical that a total reconfiguration of your IT is the solution? This is Stage Three, where what SAP calls ‘collaborative business maps’ come into play. Large suites like SAP’s have a huge range of options and while no-one is going to want them all, if you’re moving to make-to-order you will need APS to check capacity on the fly. You will also need to collaborate with key suppliers to see availability. All of these will need new working practices. SAP offers what amounts to a best practices blueprint, pre-configured software with examples. This provides detailed views of the end-to-end processes and helps define activities, roles, interfaces and the impact of change. Others offer this too: JD Edwards, for example, provides a Solutions Modeller, an interactive graphical modelling tool that allows companies to define and validate processes. And Geac has similar systems. The value is in helping to prevent missing steps, and learning from others. Even if you get it right initially, though, what’s to keep you on the virtuous path? The change process is so demanding that a reluctance to tinker with the end-product is understandable. But there’s only one word for companies who don’t go on: threatened! Some use formal benchmarking, others use automated solutions: SAP can pull key performance indicators (KPIs) from its system into a data warehouse where imported metrics can be compared. But you don’t have to be this sophisticated. Any company with a handle on its processes can do it. In fact, sharing information on information boards across a factory can often reap better rewards than flashing it across managers’ screens. Continuous questioning is the only guarantee of continued profitability.