Generation of test method boundaries for refrigeration technology

When manufacturing refrigeration appliances, it makes sense to minimize the number of defective products. This can be achieved in several ways. Since the causes of marriage are great, only those that are determined by a person and are not unambiguous should be considered. One of the possible reasons for its decrease is the accurate determination of the numerical values of the parameters of a suitable refrigerator. A deviation from the values obtained in this way will form a suitability range for each temperature. The task is facilitated by the presence of two numerical parameters for suitability: temperature and power. Consequently, if the power is at the level of the working standard refrigerator, and the temperature gained over a certain time is higher than it should be, then the refrigerator can be unambiguously recognized as defective. This article discusses the possibility of an automated calculation of the boundaries of the parameters of the acceptance test methods (ATM). They are carried out before the release of the finished product to the consumer. The calculation is based on real production statistics. The test data forms parameterized point clouds for power and temperature. Classifying points in the cloud will make the boundary calculations more accurate. This will lead to a decrease in the human factor when deciding on suitability. Consequently, the accuracy of determining the defective product will increase.