Release Date: July 2004

MG Rover User Story

Improving vehicle build quality at MG Rover Group with Icona Solutions’ aesthetica™ production variability visualization software.

In the automotive industry, perhaps more than in any other and in particular when it comes to the private car sector, design is becoming one of the most important differentiating factors in the customer’s selection process. An eye-pleasing design that reflects the corporate image of the manufacturer and the vehicle’s purpose is what attracts a potential customer’s attention. But what retains it beyond that initial attraction is the apparent quality of the vehicle in question.

When it comes to the customer-visible surfaces of a vehicle – the external body panels, the instrument panel, the interior trim and the different interior and exterior components and assemblies – the assumption is that if these appear to be of a high quality, in terms of finish and the way they fit together, then the usually invisible parts of the vehicle are probably of a similarly high quality.

So as well as good design, visibly high build quality is an essential element in the car buying process today. As a result, achieving that quality, consistently, is something that is high on the list of requirements of all OEMs. Nevertheless, satisfying the requirement - affordably - is something else.

One automotive company that is actively working at becoming an industry leader in achieving consistently high build quality, without adversely affecting its manufacturing costs, is MG Rover Group.

An independent, mid-sized and British-owned company with some 6,500 employees, MG Rover Group was formed in May 2000 when it was acquired by its present owners, The Phoenix Group, from BMW. From its Longbridge International Headquarters in Birmingham , at the core of the UK ’s engineering and manufacturing heartlands, the company designs, engineers, manufactures and markets a wide range of models and body styles under the Rover and MG brand names.

The Rover line-up comprises the CityRover and Streetwise city cars, the 25 and 45 series of compact family saloons and hatchbacks and the Rover 75 and 75 Tourer medium family saloons and estate cars, as well as the Commerce cdv (car derived van). Meanwhile, the sporting MG range comprises the ZR, ZS and ZT sports saloons and hatchbacks, the ZT-T sports tourer, the MG-TF two-seat sports car and the Express cdv. Topping out the MG range is the MG XPower SV, a 400 bhp, V8, high performance two-seat coupe.

The drive for quality.

As is the case at all automotive companies today, the designers and engineers at MG Rover rely on the latest CAD/CAM and other associated engineering software tools to help them in their day-to-day work. In the case of MG Rover, Catia from IBM/Dassault Systemes is the principal engineering design and manufacturing system, with Alias Studio Tools being used for early concept design and ICEM Surf for Class A surfaces design. In addition, a range of computer-aided engineering tools is used for such tasks as structural and finite element analysis, tolerance stack-up and tolerance analysis etc.

All of these tools taken together go towards helping to ensure that designs are fit for purpose, that they will assemble correctly and that they can be cost-effectively manufactured.

Of course, as with all manufactured products, a degree of tolerance has to be included in the design of every component and assembly to take account of the variability that will occur during the manufacturing and assembly processes.

And it’s here that problems with achieving the required consistent, final build quality arise. But this isn’t a new problem. Nor is it one that is peculiar to the automotive industry. Tolerance allowance, tolerance stack-up and production variability are disciplines that are well understood by most design and manufacturing engineers.

However, it is a problem that is hard to overcome when it comes to ‘cosmetic’ components and the aesthetics of a product. Then, finding the answer to a problem becomes more than purely a question of mathematics. Subjectivity also comes into play.

It is here that MG Rover is probably at the forefront of applying the latest software technologies to assessing the effects of production variability on aesthetics – through the use of Icona Solutions’ aesthetica™ software.

As Alan Olifent, manager of dimensional quality in MG Rover’s concept integration department, explains, “The secret of quality is not that you know that things will vary dimensionally – because they inevitably will – but that you know by how much they will vary and that you have taken that into account in the design and manufacturing processes.”

“If you know that the level of variability that you’re likely to experience in the manufacturing and assembly processes is smaller than the variability that you’ve protected against in the design phase, then you’re not going to experience any nasty shocks in the final assembly phase. The question is though, does the final product meet your aesthetic quality targets? To achieve this, you need to know that at the limits of the allowable variation, the product is still acceptable aesthetically. And here it isn’t a black or white decision”

This is the problem that Olifent and his team at MG Rover recently set out to overcome, with the help of Icona Solutions.

Tolerance and variability.

Using today’s CAD/CAM and visualization software technologies, designers and engineers are able to produce convincing, photo-realistic images of products before anything has been manufactured. These design visualisations are used routinely throughout the design development process for design reviews and for customer and senior management presentations.

The only trouble with them, from an engineering perspective, is that they are generated from the nominal dimensional values. They show the vehicle or assembly in its perfect form. But what happens when something strays from the nominal during the manufacturing process?

Traditionally, when designers assign acceptable tolerances to nominal dimensional values in order to allow for production variability, the process tends to be in the form of a ‘best guess’ or by using information from previous projects. However, problems can arise down the line when tolerance stack-ups indicate that an assembly exceeds the overall tolerance limits, with the result that the individual components don’t fit together properly.

So MG Rover’s objective was to protect against dimensional and positional variation - in the design phase of a vehicle programme. Their aim was to be in a position in which they are able to assess the effects of tolerances and production variability on the aesthetics of an assembly, or a complete vehicle, so that engineering specifications and manufacturing processes can be matched beforehand in order to achieve higher yields and quality.

As Olifent explains, “It’s about clarifying the engineering specifications in the first place, then interrogating them to ensure that the allowable limits are acceptable aesthetically. If they are not, you can make the necessary changes in the design phase and avoid the need for costly downstream changes later.”

While traditional tolerance stacking is a relatively straightforward process, it becomes more difficult to assess the effects of tolerances and production variability when it comes to the 3D design world. Added to this is the fact that the effects of variability can come from areas of a vehicle other than the assembly that you’re measuring.

For example, during the analysis of the front-end alignment of one model in the Rover range, one of the first variables that needed to be considered was a tooling hole at the rear of the vehicle. This forms one of two datum points for the body mounting, the other being near the front of the vehicle. So any variability at the rear tooling hole immediately affects the front-end assembly. Further, any variation between the two datum points, for example rotation, also has an effect.

The result is that, to undertake a variational analysis of the front-end assembly requires a full body digital model. This immediately introduces many more sources of variability, all of which need to be understood, measured and controlled before the front-end alignment can be controlled.

Again, there is nothing particularly unusual here. Today’s tolerance analysis tools can handle situations such as this without too much difficulty – as long as you are only concerned with functional, solid parts.

Subjective acceptability.

The real problem arises when you start to include sheet metal parts, such as body skins. Sheet metal is only ever semi-rigid. Never rigid. So when it comes to assessing the effects of variability on body skins and other flexible components which are inextricably linked to the aesthetics of a vehicle, you start to enter the realms of subjective acceptability.

As MG Rover’s Olifent puts it, “It’s all very well applying a tight tolerance in order to reach a desired level of quality – but it may not be achievable in reality. If it isn’t, the question then becomes, ‘By how much do we miss?’ One answer to the problem might be to increase the allowable tolerances – but you then get a poor quality product.”

Of course, it is always possible to iron out some of the effects of production variation during the final vehicle checking process, for example by adjusting the hinges or latches to bring the door shut lines back within the acceptable tolerance limits. But there’s a cost attached to this. So a cost/benefit analysis is required, which involves a trade-off between allowable variation and right-first-time production.

But when it comes to the more cosmetically sensitive areas of a vehicle, the big question here is ‘What will it actually look like?’ And that’s a question that needs to be answered before problems begin to appear on the production floor.

And it’s the one that MG Rover has set out to answer with aesthetica™ . Because, while currently available tolerance and variation analysis tools go some way to providing the required information, they don’t go far enough. A complex report comprising a series of histograms is fine for the engineers that will interpret them. But it doesn’t show anybody what the assembly will actually look like.

That’s where the use of aesthetica™ comes in at MG Rover.

An example of the software’s ability to enable people to see the effects of variability on the aesthetics of a vehicle occurred on an earlier MG Rover vehicle programme. Here, one of the biggest issues was a potential problem with wind noise. Within the problem area, the design called for the upper half of the front door and the A pillar to be flush – but in practice, there was still a problem with wind noise. Analysis of the area indicated that the problem would be solved if the door was under-flush with the A pillar.

However, the designers and the engineers could not agree on the solution. Without going back to the CAD model and changing it – which was out of the question at that point – the designers couldn’t see what the effect of the changes would be on the vehicle aesthetics. They therefore couldn’t agree to the change. So after several weeks the problem still hadn’t been resolved.

So with aesthetica™ , Olifent used the analysis data, in conjunction with the CATIA engineering CAD data, both of which were imported into the aesthetica™ environment, to create a visualization of the effects of the changes suggested by engineering. He then gathered the heads of engineering, design and body-in-white together in MG Rover’s visualization suite and showed them what it would actually look like in reality. The decision to make the changes was made within minutes – and the wind noise problem was solved.

But that was solving a problem on an existing vehicle. On future MG Rover vehicle development programmes, aesthetica™ will be used in the early design phase to help in setting aspirational tolerance – and thereby, quality – targets against which the design can be measured as it evolves. Further down the development process, it will also be used to visualize the effects of anticipated production variability on the aesthetic quality of the vehicle in question so that any potential problems can be ironed out before production commences.

By doing this, MG Rover Group will have gone a long way to achieving its aim of consistently high build quality in all its vehicles. And of retaining the attention of a discerning buying public.