Modeling Odor Optimization of Vehicles Based on Data-Driven Goal Programming

In recent years, there is an increase of customers’ requirements on the comfort of their vehicles. As a result, reducing the odor inside the vehicles has become an important and elusive task. Extensive experimental results show that the odor inside vehicles mainly comes from VOC (volatile organic compounds) emitted by the interior ornaments and parts. Given there are many VOC components affecting the odor, determining which VOC components are essential to the odor becomes a main problem in optimizing the odor in vehicles. In our paper, we propose a new approach to optimize the odor of VOC in vehicles based on data-driven modeling and goal programming. To this end, we first collect mass spectrograms of the interior ornaments and parts and their odor ratings, where the mass spectrograms are obtained by IMR-MS and ratings are rated by olfactory engineers right after the specimens undergo bag VOC tests. Then we use those data to build a data-driven model based on Weber-Fechner Law. The data-driven model is solved using lasso regression. Based on the data-driven model, we find out the contributions of the VOC components to the odor rating, which enables us to focus on certain specific VOC components that contribute much to the odor ratings. By strategically reducing those specific VOC components using goal programming, we finally obtain an optimized design with a better odor rating. To be specific, when performing the optimization, instead of minimizing the VOC odor rating, we set an ideal odor rating as the goal and formulate the optimization as a goal programming problem. We further consider the tractability of reducing the VOC components on the selected mass weights and assign different weighting factors to those VOC components based on the easiness of reducing them, which makes the finalized design of much practical significance. To validate our approach, we collect 179 VOC mass spectrograms to train and test our data-driven model. The average accuracy of predicting odor ratings from mass spectrograms can reach 85% ~ 90%. Using this data-driven model, we select the VOC components over several mass weights that have high contributions to odor ratings. The physical properties of the selected VOC components show that they are irritant and volatile, proving that our selection is reasonable. In conclusion, we propose a novel approach that has two contributions: 1) it can accurately reflect the relationship between VOC mass spectrograms and odor ratings of the interior ornaments and parts in vehicles; 2) it can find a proper way of selecting and reducing certain VOC components to reach a target odor rating. We believe our approach will be useful to the researchers and engineers in the vehicle industry.