Mastitis detection and prediction of milk composition using gas sensor and electrical conductivity

Abstract

Milk is a complex raw material, and it requires a strict quality control. The analyses that milk undergoes upon its arrival to the dairy industry are essential for its quality control. However, some of these analyses are complex; expensive; and in some cases, subjective, such as the detection of sub-clinical mastitis. With that in mind, a portable device was developed with the objective to create a fast, accurate, and easy-to-use tool for the detection of mastitis and for the evaluation of the overall quality of milk. Samples of bovine raw milk were evaluated for acidity, composition, somatic cell count (SCC), and electrical conductivity, coupled with gas sensor MQ-135, responsible for detecting carbon dioxide and ammonium, and gas sensor MQ-3, responsible for detecting ethanol, benzene, methane, hexane, and carbon monoxide. Through the MQ-135 measurement, it was possible to determine the occurrence of mastitis milk with a 77% accuracy, while MQ-3 did not show promising results. Besides, it was also possible to estimate acidity, lactose, protein, ashes, and casein content with the use of our portable device, with a satisfactory level of accuracy when milk components were found within their normal range of variation. The sensor device developed shows potential to provide a fast decision-making tool in the dairy industry.