摘要：Deterministic and probabilistic models of measured quantities, processes and fields in control systems of manufacturing processes, as well as physical and probabilistic measures, make it possible to form a measurement result, to provide it with the properties of objectivity and reliability. The measuring instruments necessary for obtaining new knowledge on the manufacturing process are being developed and improved. Issues of improving and developing models and measures in measurement methodology play an increasingly important role in achieving high measurement accuracy and expanding the areas in control systems and expert systems for manufacturing processes. The measure is formed by many factors, the action of most of which is of a random nature. It is possible to determine such a measure as probabilistic measure, which can be applied, for example, transmission of measurement data via communication channels, registration of the measurement results of manufacturing processes. The stochastic approach in the theory of measurements is of particular importance in the case of measurements of physical quantities that have a probabilistic nature. The basic idea of the paper is to do transformation from the well-known triad "model → algorithm → program" to a more reasonable methodology "model → measure → algorithm → program". The methodology allows to improve accuracy and reliability of the obtained measurement results of control systems and expert systems of manufacturing processes.
关键词：statistical methods;reliable measurements;errors in variables identification;signal models;random field models;probabilistic measure
