Academic Facets: News and Views from Spatial's University Community

[#1]

I just spent a day at a vehicle manufacture in the West Midlandsarea of England where  I’d been invited to speak at an associationof manufacturing engineers (mainly from the automotive supply chain) who holdquarterly meetings at factories throughout Britain. In good times the groupacts as a supports network to share information about new technologies andinsights to common problems, but this meeting provide more moral support thantechnical discussion. Everyone’s story was the same, October was a “normal”month, orders and production were good, but in November “it was like someoneturned out the lights”. Nearly all commercial activity appeared to havestopped. The factory we were meeting in had just started an extended five weekshut down over Christmas. The situation would be “assessed in the New Year”,said the production director with a shrug. 

However despite the inactivity our factory tour went a headand because everything was idle we could actually hear our guide (a hugeimprovement on the average walk round). The “Mary Celeste” state of the plantalso meant we could walk freely around the plant with out fear of big squashedby forklifts, hit or welded. Although I’ve seen it many times before, I neverfail to be impressed by the astonishing complexity of modern manufacturing, thetools (aka dies or moulds) required to press out the body panels, for example,are works-of-art whose production requires complex surfaces, 5-axis machiningand CMM inspection . We passed through a warehouse that was stack from floor toceiling with such tools sets for making body panels for vehicles no longerproduced: “We still use them for making spare parts” said our guide.

I hope the company I visited will survive, (it has a richowner with deep pockets who can afford to wait out the storm) but others willnot be so lucky. It appears likely that across the world hundreds of manufactureswill become insolvent and their buildings and equipment put up for sale.  But one person’s distress is often another’sopportunity.

There is a long tradition of small niche manufactures buyingup the tooling of bankrupt enterprises and making a living from a near monopolyon particular spare parts. If you are the only person in the world able to makethe side panel for a 1958 Sports car then you can, almost, name your price.

But I wonder how much value is put on the CAD/CAM data? Sureif a product is going to continue in production then the CAD data would beassociated with the IP. But what if it is literally the end of the road for aenterprise and its products, does the CADCAM data simply get trashed with thecomputers? Some where in every engineer organisation will be many Mbytes of 3D componentmodels that provide a time line of a parts slow progression from prototype to productioncomponent.  

Anyone (any academic at least) involved in geometricmodelling research knows that it is extraordinarily difficult to get hold ofreal components for testing/benchmarking new algorithms against. The NationalDesign Repository at Drexel University addressed thisproblem in the 1990’s but always lacked a good supply of parts with complexsurfaces.  So although it helped boostwork on 2.5D geometry it had little impact on the development of methods forprocessing the complex surfaces that now dominate the CAD/CAM world.

In other words if you are a researcher with a hot new way ofdoing Booleans or identifying features, or generating tool paths, or modifyingsurfaces, or calculating spring back, or generating a process plan for partswith B-spline surfaces you frequently have to create your own test parts. Andexperience shows that ones own test parts only ever incorporate the problemsyou have thought about, never the ones that hadn't occurred to you! Indeed, whilethe problem is significant for purely shape based algorithms, its is absolutelychronic for people studying the evolution of product designs and the organisationof engineering enterprises, where access to a complete electronic paper trail(ie models, emails, memos, PLM database etc) is unheard of.

Someone should be asking the research funding bodies formoney to selectively purchase the CAD/CAM/PLM data of defunct companies, andmake it freely available on the web in a structured coherent way for CAD/CAMresearchers to use.  Hmmmmm where is thatapplication form?