Research On Control Chart Of Autocorrelated Process Based On “GO/NO GO” Ring Gage

Since the founding of New China,especially since the reform and opening up,China's manufacturing industry has developed rapidly,which has effectively promoted China's industrialization and modernization process.With the introduction of concepts such as "Industry 4.0" and "Made in China 2025",countries are paying more and more attention to manufacturing,and the quality of their products has received much attention.In the production process,the basis for judging whether the quality of the product meets the standard is whether the quality characteristic data of the product is within a reasonable range.For some highly standardized and simple and easy-to-process products,mass production is usually used.During the inspection process,some parts,such as hole shafts,are small in size.Measuring the quality characteristics of these parts is very time-consuming and easily affected by various aspects which results in inaccuracy,so “GO/NO GO” ring gage was invented for these production processes.“GO/NO GO” ring gage is divided into “GO” ring gage and “NO GO” ring gage.The upper limit of the product's allowable deviation is used as the stop side,and the lower limit of the allowable deviation is used as the pass side.During inspection,products are qualified if they pass “GO” ring gage and doesn't pass “NO GO” ring gage,If they pass all ring gages or don't pass all ring gages at the same time,they will be disqualified.Due to the simplicity of operation and high detection efficiency,“GO/NO GO” ring gage is widely used in quality inspection.Traditional control charts assume that statistical data exists independently.However,in some production processes that use “GO/NO GO” ring gage to implement inspection,including mass production of parts forging,hole and shaft extrusion,with the development of measurement technology and data collection technology,the assumption of data independence has been questioned.Autocorrelated data was found in these production processes.For these production processes with autocorrelated data,if the conventional control chart is used for quality monitoring,a large number of false or missing alarms may be generated,which will influence the monitoring efficiency and bother the order of the production process.It alsoleads to unnecessary quality control costs.For the autocorrelation phenomenon in the production process using the “GO/NO GO”ring gage to detect,this paper proposes a control chartxnp AR-control chart to monitor the mean value of the autocorrelated data.Because of adopting “GO/NO GO” ring gage to inspect,this control chart takes full advantage of simple operation and high monitoring efficiency.In order to optimize the design of this control chart,compared with the conventional Markov chain method and Monte Carlo stochastic simulation method,this paper uses a easier method the Stein-chen method to calculate the average run length(ARL).In addition,comparing the performance of the proposed control chart with the traditional attribute control chart,it shows the superiority of the control chart in processing autocorrelated data.This paper also provides a sensitivity analysis to study the robustness of the proposed control chart,and finds that thexnp AR-control chart has good robustness when the mean shift is small and the correlation is low.Finally,an industry case is used to illustrate how to apply the control chart to the production process,which provides a reference for practical applications.