Data-Driven Machine Learning Approach for Rattle Risk Mitigation

In the modern automotive industry, squeak and rattle issues are critical factors affecting vehicle perceived quality and customer satisfaction. Traditional approaches to predicting and mitigating these problems heavily rely on physical testing and simulation technologies, which can be time-consuming and resource-intensive, especially for larger models. In this study, we propose a data-driven machine learning approach to mitigate rattle risks more efficiently. In this study, a floor console model has been evaluated using the traditional simulation-based E-line method to pinpoint high-risk areas. Data generation is performed by varying material properties, thickness, and flexible connection stiffness using the Modified Extensible Lattice Sequence (MELS) sampling algorithm, creating a diverse and comprehensive dataset for generating a machine learning (ML) model. Utilizing the dataset, the top contributing variables were identified for training the ML models. Various machine-learning models were developed and evaluated, and the best-performing model was selected based on accuracy and generalizability. A Genetic Algorithm (GA) was employed to optimize the system further, in conjunction with the selected ML model to determine the optimal set of design parameters for rattle mitigation. The optimal operating parameters were validated with simulation results confirming the model's reliability. This optimization process significantly outperformed traditional methods, yielding a time gain of 92 times compared to the solver-based optimization approach with a similar level of accuracy. The proposed methodology reduces computational time along with providing a robust framework for efficiently mitigating rattle risks, highlighting the potential of machine learning and data-driven optimization in engineering applications.