One of our main research lines focuses on the development of expert systems aiding product design. Expert systems allow to use knowledge of a particular topic in decision making. In the field of design, they are used to simulate the process by which certain reactions of users interacting with products are analyzed and design features are decide based on the results of such analysis.

Generative design

Generative design is one of the pillars we use in the development of expert systems. This concept is based on defining a parameterization of the object to design suitable to the design objectives. The parameterization generates an abstraction of the object which allows producing numerous alternatives simply varying the values ​​of the defining parameters. Generative design uses this parameterization combined with some type of algorithmic procedure, for example genetic algorithms, for creating alternatives which follow a preset target.

Neural networks

Artificial neural networks (ANN) constitute our other key pillar when developing expert systems. The RNA are models that replicate the performance of the biological nervous system. Its goal is to “learn”, in a similar way to how the brain does, in the context of a given problem. The use of RNA can, with proper training, get “user models” networks whose response to product stimuli emulates that of the consumer. The implementation of networks in combination with generative design allows the immediate introduction of user judgment regarding any product alternative.