Back to news

Folha: Bézier was a pioneer of mathematics in car design.

Foto: Wikimedia Commons

Reproduction of Marcelo Viana's column in Folha de S. Paulo.

In 1933, the young Frenchman Pierre Bézier (1910–1999) obtained his first job at Renault, as a tool installer. He retired from the same company 42 years later as director of mechanical methods. In the meantime, he revolutionized automation in the design and production of Renault and Peugeot cars, leaving an extraordinary legacy as a pioneer of computer-aided design (CAD), which today permeates countless sectors of activity, from industry to the arts.

Bézier graduated as a mechanical engineer from the École Normale Supérieure des Arts et Métiers in 1930, and as an electrical engineer from the École Supérieure de Électricidade the following year. Much later, at the age of 67, he enrolled at the Pierre and Marie Curie University to obtain a doctorate in mathematics. Alongside his work in industry, he was a professor at the Conservatoire National des Arts et Métiers in Paris between 1968 and 1979.

Read more: Summer Course opens applications for scholarships
Faperj Cycle promotes lectures with IMPA researchers.
IMPA is hiring a project scientist.

One of the biggest problems in the automotive industry was how to model the shape of a new car so that it would be both aerodynamically efficient and aesthetically pleasing. One possibility would be to consider various equations, have the computer draw the corresponding surfaces, discard those that were unsatisfactory, and refine the promising ones. But it's not that simple.

To begin with, it would be necessary to discover reasonable equations for the surface area of a car: besides being laborious, such equations would certainly be very complicated. This was even more critical at a time when computers were in their infancy: in 1964, after much insistence, Bézier managed to get Renault to acquire a second-hand computer with an incredible eight kilobytes of memory!

Furthermore, even if an equation yielded a promising result, it wouldn't be obvious how to modify it to improve the outcome. In any case, it was counterproductive to spend time and money finding and improving complex equations that, frankly, served no other purpose.

It would be much better if the computer could draw different shapes without needing equations, and in a way that these shapes could be easily adjusted and modified by the user to obtain even better results.

To read the full text, visit the newspaper's website.

Read also: IMPA has openings for researchers and postdoctoral fellows.
OBMEP: PIC students visit federal universities