A radical new approach to vehicle manufacturing

“Rather than basing the manufacturing facility on a single model line based design, we’ve simulated a multi-model matrix of cells for car manufacturing,” says Warwick Manufacturing Group’s (WMG) senior researcher, Ian Pearson. “This is a radical concept for the car industry, despite being common in other areas of manufacturing. The benefits stem from the ability to take individual cells out of production and modify them to suit current needs. Extra cells can be added or deleted to account for variations in sales volumes, which leads to higher utilisation of the overall facility.” It sounds very convincing, but how would it work in practice? We need to consider the car industry itself. First, customer demand for cars is extremely volatile. Add to this the issues of global overcapacity, third world competition and new legislation constantly driving changes in vehicle design and production, and you’ve got a real business conundrum. And when manufacturers design new car models, they typically need to invest millions of pounds in new plant and production line equipment. But the real problems start when the car doesn’t sell as well as originally forecasted. If this happens, your new plant investment begins to look like a bad idea, and the result is poorly utilised plant. Data from the SMMT (Society of Motor Manufacturers and Traders) annual production statistics report in 2000 reveals that most of the major car-makers operated with plant utilisation levels (on new models) that averaged around the 50% mark. This shows they were all over-optimistic in their sales forecasts, and profits suffered as a result. So, other than waiting for someone to invent accurate forecasting software, what can car manufacturers actually do to alleviate the pains of unpredictable demand? There may be an answer. Research currently being carried out by the Warwick Manufacturing Group, part of the School of Engineering within the University of Warwick, and funded by the DTI (Department of Trade and Industry), has shown that employing radical, new manufacturing methods may well be the way forward. Over the past year, WMG has been using simulation software from Delmia to test its new RAMP (responsive automotive manufacturing plant) concept. It’s already attracted a lot of attention from UK automotive manufacturers, and in the last few weeks, representatives from Lotus, Land Rover and Jaguar have visited the Warwick site to evaluate RAMP. Using Delmia’s Quest software to handle the complex variables surrounding the simulation, Pearson and his team have created a ‘virtual’ multi-model vehicle production plant that’s capable of making any car model, in any order, at low volume, profitably. Quest helps by making it clear which production cells are in the highest demand and these are then duplicated. Sub-assemblies are kitted-up onto tooling pallets and routed through the various cells. The crucial thing is, all parts of the facility, cells, tooling pallets, etc are modular and easily reconfigurable. “Bottlenecks caused by equipment breakdown no longer happen within our virtual plant,” explains Pearson. “RAMP simply re-routes within the matrix. We’ve shown that 12 different [car] models can be simultaneously manufactured by the same jigs and robots.” RAMP shows that any individual manufacturing cell can be maintained or replaced at any time without disrupting production. And this goes for upgrading equipment or modifying plant too. Equipment can be acquired and replaced as demand dictates rather than using inaccurate forecasting methods. Flexible manufacturing Ken Young, WMG researcher, says: “RAMP is relevant to Tier ones with multiple lines for different VMs [vehicle manufacturers], VMs with many niche markets, and companies with uncertain demand profiles.” And the hard benefits of RAMP? “More cost effective production of a large product range,” adds Young. “You can produce many models in the same facility with better plant utilisation. It’s flexible manufacturing really. But it’s also a phased introduction of plant, and therefore a phased investment.” The RAMP theory does look good but what about the effect on the workforce? They’ll need to be flexible and well-trained, with increased labour skills and knowledge. And that will cost time and money. And companies would need some sort of advanced scheduling to cope with the complex re-routing of parts through the matrix of cells. This will take time to implement. For RAMP to work as envisaged, firms will also need to consider the effect on product design. Design for Assembly and Design for Facility techniques will have to be embraced for it all to succeed. Nevertheless, “plants that can cost effectively produce different models and change from an old model to a new one quickly will have the competitive advantage,” observes Young.