Reliable measurements are critical for the food industry. The measurement uncertainty assessment is an obligatory procedure stipulated in modern standards, e. g., general requirements for the competence of testing and calibration laboratories (ISO/IEC 17025); Guide to the Expression of Uncertainty in Measurement (GUM), etc. This research applied the hydrometer and pycnometer techniques to the uncertainty associated with milk density tests. The tests employed both hydrometer and pycnometer methods. Measurement uncertainty was calculated using the simulationbased approach and uncertainty distribution laws. The hydrometer method revealed such uncertainty sources as thermometer calibration, hydrometer calibration, and repeatability. For the pycnometer method, they included balance calibration, balance resolution, weighing repeatability, pycnometer calibration, and repeatability. Both techniques were provided with mathematical models that incorporated input variables with assigned distribution laws, standard uncertainties for these variables, and the total relative standard uncertainties. The total standard uncertainty for milk density measurement was 31 kg/m3 for the hydrometer method and 21 kg/m3 for the pycnometer method. The expanded uncertainties were computed at a 95% confidence level with a coverage factor of two. The measurement uncertainty depended on the method and the instrumental errors. The hydrometer and pycnometer tests showed different measurement uncertainty, depending on the method and instruments employed. The pycnometer method had lower uncertainty and provided more reliable and accurate measurements. Selecting an optimal method and accurate calibration minimize uncertainty; however, additional factors require further research to adapt the models for industrial applications.
Khamkhanova et al. (Tue,) studied this question.