Statistical Methods for Assessing the Quality of Technological Processes With Limited Information

The article discusses the issue of improving the efficiency of quality management in the mechanical processing of machine-building products in small-batch production. The main focus is on the application of statistical control methods, in particular the cumulative sum method, as a tool for diagnosing the accuracy of the technological process on numerically controlled lathes. The feasibility of using cumulative sum control charts instead of traditional methods that do not take into account the dynamics of parameter changes over time is justified. The study analysed the results of mechanical processing of a crankshaft flange made of 20X steel on a 16K20F3 machine. Twenty-five samples of five products each were taken to assess deviations from the nominal value and construct the corresponding cumulative curves. Based on the experimental data, graphs of cumulative sums were constructed, and a reference chart for use in production was proposed. A clear relationship was established between the slope of the cumulative sum curve and the quality level of the machining process: a horizontal curve corresponds to a satisfactory state, a downward curve corresponds to poor quality, and an upward curve corresponds to high quality. A mathematical apparatus is proposed for calculating a comprehensive indicator of machining quality, taking into account the weight of individual quality indicators (accuracy, roughness, spatial deviations, etc.). It has been established that this comprehensive indicator correlates with the level of the technological process and can be an effective criterion for its assessment. The results of the study can be used for operational control and quality improvement in the production of small batches of machine-building products, which is especially relevant for small-batch and multi-product production.