Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Endangered engineers

Share this on social media:

Nick Morris talks to Chris Randles, CEO of Mathsoft, and Alan Stevens, an engineer from Rolls-Royce, to examine the state of the mathematical modelling sector

Engineers are fast becoming an endangered species. The retirement of the baby boom generation is imminent - more than half the engineers currently working are over the age of 45 - and the number of engineers currently being trained in universities on both sides of the Atlantic is not high enough to meet the demands of industry. Larger firms are increasingly aware that it is becoming difficult to find enough young talent to feed market expansion.

Chris Randles, CEO of Mathsoft, is acutely aware that the shortage of engineers will generate a variety of problems for industry in the near future. In an interview during a recent visit to London, he cited examples of major engineering projects whose lifetime may extend beyond the career of a single generation of engineers.

For instance, the Boeing 747 started life as a concept in 1963, and took its first flight in February 1969. The latest development of the same aircraft took to the air in 2002, and will see many years of service yet. Another case is that of the F-15 fighter jet, a project begun in the late 1960s, which is also expected to see active service into the second decade of the 21st century, well beyond the retirement of even the freshest faced engineer on the original research team. The design of a fuel filler cap for the F-15 alone generated 75 pages of calculations and other related information. Expand this across the whole aircraft industry and you get some idea of the sheer quantity of knowledge that somehow must be passed from one generation of engineers to the next to ensure effective maintenance, development, and above all, continued safety of the aircraft.

According to Randles, Mathsoft's Calculation Management Suite attempts to make the joint issues of knowledge capture, retention, and transfer, easier to manage, by moving computer-based design from an individualist, point operation, to a server-based group approach to calculation and data storage. Under the new system, data stored on a central server, 'the design DNA', can be accessed easily by any number of engineers - making group projects much easier to manage. The calculation data for any process is also stored and easily accessible, enhancing design traceability on both the solo and collective level. Because more people can access, examine, and re-trace calculations, errors are more likely to be spotted and rectified at an earlier stage in the product life cycle.

However, the real world is seldom as tidy as this. Different individuals within a workgroup, or different groups within an organisation tend to use different mathematical software tools for design and development. Each group will use the tool best suited to the particular work being carried out. Each design group may well operate on its own private island in the corporate sea, using whichever mathematical software product best meets its own requirements. Unfortunately, data doesn't always transfer perfectly between different applications.

Mathsoft is addressing the problem of interoperability through partnerships with other companies that supply maths tools, and its membership, alongside Maplesoft, National Instruments, and the French INRIA - all members of the Numerical Mathematics Consortium, an organisation that aims, among other things, to develop standardised programming algorithms and semantics. By undertaking software development with these partners, Mathsoft can ensure that data and more complicated calculations can be cleanly transferred from one platform to another, and that those different results will give the same answers to mathematical problems.

The work of Mathsoft, and other software companies, encourages engineers to use dedicated mathematical software for design purposes. Unfortunately old habits die hard. In large companies, such as British engineering firm Rolls-Royce, the software package common to all the desktop machines is the ubiquitous MS Office suite, with its associated spreadsheet application Excel. Alan Stevens, a specialist in mathematical modelling and simulation at Rolls-Royce, noted in an interview: 'Many engineers will use a spreadsheet for quite complicated design calculations - just because it's there.'

The use of spreadsheets can cause headaches for someone trying to trace a design fault. In some cases, such as in a nuclear reactor, such a fault could not only be costly, but dangerous. Although it might be easy to find and access the original design spreadsheet, understanding the information it contains can prove more difficult. The way the cells in a sheet are organised, with cryptic data labels, can be obscure at the best of times - according to Stevens, the person who compiled the original spreadsheet can often have trouble retracing the various calculation steps, even the next day!

This is a live issue for Rolls-Royce, which, as the design authority for the Royal Navy's naval nuclear plant, designs, supplies and supports all the pressurised water reactor (PWR) systems and equipment that power the UK's submarine flotilla. Rolls-Royce is also supplying the reactor plants for the Royal Navy's next generation of nuclear submarines - the Astute class, for which two more submarines are planned.

But the job is not finished once the submarines are launched. There is maintenance and continuing safety monitoring. In 2001, the company also signed long-term contracts worth up to £760m with the UK Ministry of Defence to ensure the safety, availability and performance of the Royal Navy's submarine flotilla.

Stevens uses mathematical software as part of his everyday design and calculation work for the company. He cited one piece of work in which he's independently verifying the calculations of a specialist risk assessment code called FaultTree+. In particular, he is verifying its Markov Chain modelling component prior to the company's probabilistic risk assessment specialists using it for real. He is keen to extol the virtues of bespoke rather than generic software to others in his field. As he points out, a calculation on a Mathcad worksheet looks very similar to the same thing written out by hand. Contrast this with a line of brackets, stars and coordinates familiar to anyone who has tried to carry out a non-trivial calculation using spreadsheet software. Not only is the specialist approach easier to understand intuitively, but it is also easier to spot errors. This means that, when an engineer wants to re-examine the work of another, he doesn't have to spend his time working through the maze of cells trying to understand the relationships between them; he can concentrate on the maths.