Luis F. Alvarez
SAP Logistics Consultant
Blog Post

Design Optimization based on Genetic Algorithms

In 2002 Jaguar Racing started a major redesign of its F1 car due to championship results falling short of Ford's expectations. The main objective was to reduce the overall weight of the car.

The optimization of the front wing was specially challenging after an industry review revealed that no commercial software was available to find the optimum combination of fiber orientation and number of plies that produce the maximum structural stiffness at the lowest mass.

A multidisciplinary team involving Altair Altair Engineering and the University of Bradford assessed the problem and developed an innovative solution based on evolutionary computing that was later used in the wing lay-up of the R3 formula one car.

The problem

The design of Formula One cars takes approximately 4 months and has to go through as many iterations as possible in an effort to optimize every aspect of the car. In such a competitive environment, Jaguar Racing wanted to reduce weight in the R3 composite front wing . A robust technology was needed to optimize an already highly optimized design.

The methodology

A specific genetic algorithm was developed running on Altair FEM software OptiStruct in order to solve concurrently the carbon fibre orientations and the number of plies of the wing composite lay-up.

The optimization problem was stated as to minimize the mass subject to FIA standards and aerodynamic loading.

Evolutionary Computing studies how computer programmes can simulate processes found in nature to assist in optimization tasks. The main principles are:

  • Maintain a population of solutions.
  • Measure the fitness of individuals at competing with each other.
  • Evolve according to Darwin's natural selection rules to increase quality.
  • Explore diversity with sexual operators such as crossover or mutation.
  • Breed the population over generations.
The results

Initial results showed trends of the wing lay-up (e.g. biased more to bending in the middle of the wing and plies biased more to twist at the outer edge). Final results showed up a new and innovative direction for the wing lay-up and a variation was put onto the latest model of the R3 car.

GA achieved 5% reduction over the baseline weight of the wing. Looking at only designable areas, GA achieved 15% weight reduction.

Read more