Page 168 - Visions of the Future Chemistry and Life Science

152 P. J. KOLSTON

entirely new tasks with efficiencies that are impossible using current tech-
nologies. Given the impressive specifications of such organs, these new
devices – manufactured in carbon or silicon – could have numerous
research, clinical and industrial applications.

9.2 Making cars
Computer modelling has been used extensively in manufacturing indus-
tries for many years. One familiar application is in the design of car crash-
worthiness. Cars must both protect the occupants from physical intrusions
into the passenger compartment, and minimise the deceleration forces
that act upon them. The first of these requirements could be achieved
easily by making the car body rigid. Unfortunately, the deceleration forces
would then be intolerably large, so instead the design aim is to make those
parts of the vehicle that are outside the passenger compartment absorb as
much of the impact energy as possible, by making them deform in the pre-
defined time-dependent manner that minimizes peak deceleration levels.
In the past when car crashworthiness was designed entirely experi-
mentally, full-sized prototypes were subjected to the crash scenarios
required by the relevant authorities. If the performance was unacceptable,
the shape deformations of the components making up the prototype were
examined. A new prototype was engineered empirically to overcome the
identified weaknesses before being built and then destroyed in a subse-
quent test. These tests would be repeated many times before an appropri-
ate design was found. The cost of the process was enormous.
Nowadays, car manufacturers cannot afford to destroy thousands of
prototypes when designing crashworthiness into their vehicles. Instead,
they spend most of their time building and analysing models on comput-
ers, using a technique known as finite-element analysis (Figure 9.1). Only
once a computer model is found to be consistent with statutory require-
ments do they resort to expensive and time-consuming physical testing.
This beneficial relationship between modelling and experimentation is
still in its infancy in biological research, thanks partly to the great com-
plexity of biological organs.