Predicting Friction System Performance with Symbolic Regression and Genetic Programming with Factor Variables
This work addresses a domain-specific problem for engineers designing friction systems, offering an incremental improvement in handling nominal variables in symbolic regression.
The paper tackled the problem of predicting friction system performance, which is difficult due to many parameters, by using symbolic regression and genetic programming with factor variables to handle nominal variables, resulting in models with predictive accuracy comparable to artificial neural networks and lower complexity than those using one-hot encoding.
Friction systems are mechanical systems wherein friction is used for force transmission (e.g. mechanical braking systems or automatic gearboxes). For finding optimal and safe design parameters, engineers have to predict friction system performance. This is especially difficult in real-world applications, because it is affected by many parameters. We have used symbolic regression and genetic programming for finding accurate and trustworthy prediction models for this task. However, it is not straight-forward how nominal variables can be included. In particular, a one-hot-encoding is unsatisfactory because genetic programming tends to remove such indicator variables. We have therefore used so-called factor variables for representing nominal variables in symbolic regression models. Our results show that GP is able to produce symbolic regression models for predicting friction performance with predictive accuracy that is comparable to artificial neural networks. The symbolic regression models with factor variables are less complex than models using a one-hot encoding.