Abstract: Five commercially available plant-based milk samples were selected to investigate their differences in aroma components using headspace solid-phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS). Subsequently, quantitative descriptive analysis (QDA) was performed uing an expert sensory panel to determine the correlation characteristics between the odor and QDA attributes. Moreover, penalty analysis was conducted based on consumer responses to explore priority sensory attributes for improvement and improvement directions for each product. A total of 92 volatile components were identified in the five samples, with the most diverse and abundant aroma compounds being aldehydes, alcohols, and pyrazines. These compounds imparted various nutty flavor such as peanut, bitter apricot kernel, and walnut flavors to plant-based milk. The penalty analysis results showed that blended plant-based milk did not need to be improved, while pure plant-based milk needed further improvement.

Key words: headspace solid-phase microextraction-gas chromatography-mass spectrometry; quantitative descriptive analysis; penalty analysis; plant milk; improvement of sensory attributes