aesthetica 3.3

03 August 2010

A unique offering that helps bring together the diverse team often charged with maintaining Perceived Quality. Al Dean reports

Product aesthetica 3.3
Company name Icona Solutions
Price from £17,500

Perceived Quality is increasingly on the minds of many of the World’s leading manufacturers.

The concept is simple: it’s all about how a customer feels about a product, and not about how a product measures up against set criteria. Manufacturers need to know if the product looks and feels as intended, if the way in which the constituent parts fit together create a superior looking product, if it feels like a product that its customers would want.

By assessing perceived quality, manufacturers are seeking to make a product that looks and feels of a high quality without completely over stretching costs during production.

Rendering variation in RTT DeltaGen

3D data of an Opel is loaded into aesthetica from the CAD system

Materials and texture information are added to add realism

Virtual locator scheme is either created or extracted from existing data

Tolerance assumptions assigned to a-class parts

aesthetica’s tools for variation simulation and exploration

Side by side comparison in Autodesk Showcase

The automotive industry

The automotive world is a perfect example of how improving perceived quality can be used to advantage. In this sector, there are very few highest tier manufacturers. Many marques and brands are owned by a small number of OEMs. As a result, those OEMs are looking to reduce costs across all their assets and a key to that is platform sharing. This is where a single platform (read: chassis and portions of the Powertrain) are shared between multiple brands and multiple vehicles.

A case in point is Bentley, a brand renowned for the highest quality vehicles, where fit and finish are of paramount importance and the company’s competitive advantage is founded on the experience of its customers owning a perfectly finished and engineered motorcar.

What many of its customers may not know is that the latest Bentley Continental range is actually based on the Phaeton platform, which is the cornerstone of many of vehicles from Audi and Volkswagen. By using a single platform, the Volkswagen group, which includes Audi and Bentley, is saving huge costs in terms of production and in terms of materials sourcing and design. However, the end results could not be more radically different in terms of perceived quality of the different brands.

Introduction to aesthetica

Into this environment steps Icona Solutions and its aesthetica product. I’ve been watching this system evolve for many years now – from its roots in the University of Leeds into a system that’s now in place in many of the leading design-led organisations across the planet. While it’s been heavily automotive-focussed since its inception, it’s now starting to gain a foothold in the higher- end consumer goods industry.

In short, aesthetica allows you to combine data and inputs from the two key teams that influence a product’s Perceived Quality: those concerned with manufacturing and assembly tolerances definition and the styling or industrial design team. The problem is that both teams have dramatically different skill sets, terminology and expertise.

While the manufacturing engineer is tasked with achieving the required quality during manufacturing by managing and controlling tolerance and variation, the industrial design team is concerned purely with the styling and aesthetic appearance of the product. aesthetica allows both teams to input their own knowledge and understanding and view the results in a manner that both can work with.

Without getting into too much detail, a 3D assembly model is used as the core dataset. Into this the assembly process and tolerances are added to show how the manufactured part could appear (with the given tolerances). This is then used to construct an intelligent model of the full assembly that can be photo-realistically rendered. It enables the whole team to experiment with tolerance assignments, assembly methods, materials and physical properties and see, immediately, how that will affect the end product. It can look at how extremes of variance will affect the visual quality of a product and ultimately feedback data on which to base critical tolerance and styling, surface and materials decisions.

The bottom line is if a product can be built that achieves the same quality, to much looser tolerances, then the production process, in general terms, becomes more cost effective. Let’s now look at how it works and explore some of the new capabilities that have been introduced to the system over the past 18 months – as there are some key new tools that make this more usable.

Geometry import

The starting point for the aesthetica workflow is to import the geometry files from the workhorse CAD system. The Icona team has been working on this on a consistent basis and the system now includes support for not only the standard formats you’d expect (such as IGES, STEP etc), but also has a number of new add-on options for importing data directly from Catia (V4 and V5), NX, Pro/E, Solid Edge, Inventor, etc. It’s worth noting that the architecture underneath the hood has been changed so that each translator is now componentised, so updates can be delivered to users as and when they are needed (for example, when the JT specification changes) rather than waiting for new releases of the main software. The system also now includes a number of geometry manipulation tools allowing users to work with imported data, such as the ability to mirror, copy and split components where needed. UCS shifts can also be applied in oder to reposition data from sub-systems that don’t match the core master UCS. Finally on the subject of geometry import, there’s a new tessellation engine at work (aesthetica uses mesh data created from the CAD surfaces) that gives much better, crack-free, results.

Simulation set-up

Once the 3D geometry has been brought into the system, the next stage is to define the locators for the assembly i.e. where and how the constituent parts are assembled, such as hole, slot etc. There are manual tools to do this that can be used to pick features on the geometry and match them between components, or that information can be extracted from the CAD geometry. There is also the new ability to import locator schemes, complete with tolerances, directly from existing tolerance models in other tolerance analysis applications, or from live measurement systems on the production floor.

Locator schemes are common in the automotive world and particularly in relation to platform-shared processes, the data is readily available for reuse. These come in spreadsheet format and define exactly which features are used to fix components to the underlying rigid structure, or chassis. aesthetica reads this information - the spatial position and the assembly method (hole, slot, clip etc) and variation, for example - and automatically inserts it into the model.

To attach the exterior components to the locator scheme, aesthetica simply inverts the locators, copies them to the components and forms the required constraints. Job done.

For those working in this environment, this serves several purposes. Firstly, it saves a great deal of set-up time through data reuse (it’s there, so it makes sense to use it) and secondly, it means that variant designs can be dealt with very efficiently. Once the initial model is set-up, it’s a much quicker task to load in the new A Class surface set, re-apply those locators and take care of any changes (where parts change dramatically or become redundant).

The next step is to define the tolerances assumptions for each locator. Again, this can be carried out manually or imported with the locators. Tolerance assumptions can be loaded directly from systems like QMC/ CM4D and QDAS using standard assumptions from DIN, ISO and other standards (these work on the basis of material and manufacturing process).

Pre-existing tolerance measurements and reports from your inspection department can also be loaded, allowing you to adapt real-world measurements for reuse. What these methods boil down to is the ability to simulate variance based on the data at hand, whether that’s standards-based, your own assumptions or based on prior experience. Now that the model is ready to go, what next?

Working collaboratively

The next stage is to enable all those involved in the process to work with the system in a collaborative manner. Variation can be inspected, both in a grouped mode as well as on an individual part pair basis. Most organisations will begin by finding out where the tight spots are and where tolerances have the greatest effect on the visual quality of the product, then narrow things down or adjust the locator scheme, materials, or surfaces in order to iron out the kinks.

At each stage, a photorealistic model can be viewed alongside more ‘scientific’ style display required by manufacturing engineers. The system works by using the variation in the locators as the boundary conditions for the mesh of the associated parts. Using this mesh and the tolerance data, aesthetica will deform the parts and using a virtual assembly process show exactly how it would appear. It also allows Gap and Flush conditions between components to be viewed and interacted with or any measurement points defined in an ad-hoc basis.

Documenting aesthetic quality variation

Photorealistic rendering of an Opel in RTT DeltaGen

While aesthetica includes a real-time rendering engine, there are higher-end products from third parties that present the data more photorealistically, so it makes sense to use them, particularly when it comes to power-walls, large format displays and linked systems across multiple geographies. For presentation, design review and collaboration we’re talking about RTT’s DeltaGen, Autodesk’s Showcase, Bunkspeed’s Shot and Luxion’s KeyShot.

Over the last few months Icona has built greater links to these systems to help show, in more detail, how good or bad your production ‘could’ be. Capabilities vary between each system. aesthetica can only send a single sample to KeyShot and Shot as a BIP file, but it contains the materials and camera views. For those working with Autodesk’s Showcase an FBX file needs to be output which has the benefit that multiple model variations can be loaded and then compared side-by-side. Camera views and material information can also be passed using separate files.

For RTT DeltaGen a single XML Scenario files can be output. This incorporates links to part geometry in Nvidia NBF file - standing for Nvidia Binary File) and is more intelligent than others as it includes camera views, materials, material sets, measurement points and multiple model variation in a single export. Essentially, each variation scenario can be explored quickly and intelligently as DeltaGen can switch between the variations with ease.

I don’t think I’m breaking any confidences when I say that Icona is also working with the lovely chaps at Luxology (the creators of Modo and SolidWorks PhotoView 360) on something rather special. It’s a little early to say how that’s going to pan out but there should be an importer added for Modo very soon.


I’ll be frank; I love aesthetica and its potential for organisations looking to achieve the highest quality production. It’s not often that I get to see a piece of technology that allows previously disparate teams, schooled in their own terminology and working practices, to work in harmony in a single system and find problems, make decisions on their own terms and move on with the job at hand.

I’m impressed by how the team has improved the reliability of results by allowing the reuse of manufacturing data where possible. It’s also great that the system can now work with a greater variety of third party visualisation systems, which makes huge sense in both capability and integration into existing workflows.

In short, aesthetica can help solve critical manufacturing issues that can have a real impact on how products are produced and perceived – and if you can make them without cripplingly tight tolerances but still achieve the required quality, then you’re onto a winner.

Comments on this article:

Interesting software but I don’t see how it can justify the £17,500 price. Then again, everything that’s mainly targeted to the auto market can be a little pricier considering the huge costs of making a car this becomes more moot, I guess.

Posted by Kevin De Smet on Wednesday 25 2010 at 09:11 AM

Leave a comment

Enter the word you see below: