Smart Odor Assessment - From Efficient Data Acquisition to Odor Prediction
The automated, data-driven assessment of the sensory properties of complex odor mixtures - e.g. in whisky - poses a particular challenge. As part of the Smart Odor Assessment concept, we are working on the linking of analytical and sensory data using data science tools. In addition to application-oriented and efficient analytical methods, we focus on automated data-processing and developed rule-based analytical tools for gas chromatography - mass spectrometry (GC-MS) data [1, 2]. Using American whiskey and Scotch as examples, we demonstrated this approach as well as a targeted classification of the two sample types based on sensory and analytical data [2]. Eventually, machine learning was successfully employed to predict relevant odor descriptors of whiskies based on analytical data [3]. Predicting the odor characteristics of whiskey can thus effectively support quality control and product development. The concept is to be further expanded and is already being researched in other application areas, such as plastics recycling [4]. [1] Grasskamp AT et al. 2023; https://doi.org/10.5194/jsss-12-93-2023 [2] Haug H et al. 2023; https://doi.org/10.1007/s00216-023-04883-5 [3] Singh S et al. 2024; https://doi.org/10.1038/s42004-024-01373-2 [4] Haug H et al. 2025; https://doi.org/10.1016/j.resconrec.2025.108479
- Lightning talk
Speaker
Helen Haug
Sensory Analytics & Technologies | Fraunhofer IVV
Helen Haug works as a research associate in the Department of Sensory Analytics and Technologies at Fraunhofer Institute for Process Engineering and Packaging IVV. Her focus is on odor analysis, with an emphasis on the non-food sector, such as plastics and recycling. In addition, she works on the implementation of efficient methods in odor analytics for food and non-food applications, such as the Smart Odor Assessment concept.... read more