{"history":[],"article":{"id":"cfdbde85-b87b-42af-875d-9ba5bfe31c9a","title":"Shewhart individuals control chart","slug":"shewhart-individuals-control-chart","content":":::summary\n**Name:** Individuals and moving range control chart \n**Proposer:** [Walter A. Shewhart](https://bliptext.com/articles/walter-a-shewhart) \n**Subgroupsize:** n = 1 \n**Measurementtype:** Average quality characteristic per unit \n**Qualitycharacteristictype:** [Variables data](https://bliptext.com/articles/variable-and-attribute-research) \n**Distribution:** [none](https://bliptext.com/articles/indefinite-pronoun) \n**Sizeofshift:** ≥ 1.5σ \n**Varchart:** MR chart for a paired individuals and MR chart.svg \n**Meanchart:** Individuals chart for a paired individuals and MR chart.svg \n**Meanstatistic:** xi \n:::\n\n# \nIn [statistical quality control](https://bliptext.com/articles/statistical-process-control), the **individual/moving-range chart** is a type of [control chart](https://bliptext.com/articles/control-chart) used to monitor [variables data](https://bliptext.com/articles/variable-and-attribute-research) from a [business](https://bliptext.com/articles/business-process) or [industrial process](https://bliptext.com/articles/list-of-industrial-processes) for which it is impractical to use rational subgroups.\nThe chart is necessary in the following situations:\nThe \"chart\" actually consists of a pair of charts: one, the individuals chart, displays the individual measured values; the other, the moving range chart, displays the difference from one point to the next.\nAs with other control charts, these two charts enable the user to monitor a process for shifts in the process that alter the mean or variance of the measured statistic.\n\n# Interpretation\nAs with other control charts, the individuals and moving range charts consist of points plotted with the control limits, or natural process limits.\nThese limits reflect what the process will deliver without fundamental changes.\nPoints outside of these control limits are signals indicating that the process is not operating as consistently as possible; that some assignable cause has resulted in a change in the process.\nSimilarly, runs of points on one side of the average line should also be interpreted as a signal of some change in the process.\nWhen such signals exist, action should be taken to identify and eliminate them.\nWhen no such signals are present, no changes to the process control variables (i.e. \"tampering\") are necessary or desirable.\n\n# Assumptions\nThe [normal distribution](https://bliptext.com/articles/normal-distribution) is NOT assumed nor required in the calculation of control limits.\nThus making the IndX/mR chart a very robust tool.\nThis is demonstrated by Wheeler using real-world data, and for a number of highly non-normal probability distributions.\n\n# Calculation and plotting\n\n## Calculation of moving range\nThe difference between data point, x_i, and its predecessor, x_{i-1}, is calculated as.\nFor m individual values, there are m-1 ranges.\nNext, the arithmetic mean of these values is calculated as\nIf the data are normally distributed with standard deviation \\sigma then the expected value of is, the [mean absolute difference](https://bliptext.com/articles/mean-absolute-difference) of the normal distribution.\n\n## Calculation of moving range control limit\nThe upper control limit for the range (or upper range limit) is calculated by multiplying the average of the moving range by 3.267:\n.\nThe value 3.267 is taken from the sample size-specific\nD4\nanti-biasing constant for\nn=2\n, as given in most textbooks on statistical process control (see, for example, Montgomery ).\n\n## Calculation of individuals control limits\nFirst, the average of the individual values is calculated:\n.\nNext, the upper control limit (UCL) and lower control limit (LCL) for the individual values (or upper and lower natural process limits) are calculated by adding or subtracting 2.66 times the average moving range to the process average:\n.\nThe value 2.66 is obtained by dividing 3 by the sample size-specific\nd2\nanti-biasing constant for\nn=2\n, as given in most textbooks on statistical process control (see, for example, Montgomery ).\n\n## Creation of graphs\nOnce the averages and limits are calculated, all of the individuals data are plotted serially, in the order in which they were recorded.\nTo this plot is added a line at the average value,\nx\nand lines at the\nUCL\nand\nLCL\nvalues.\nOn a separate graph, the calculated ranges\nMRi\nare plotted.\nA line is added for the average value,\nMR\nand second line is plotted for the range upper control limit (\nUCLr\n).\n\n## Analysis\nThe resulting plots are analyzed as for other control charts, using the rules that are deemed appropriate for the process and the desired level of control.\nAt the least, any points above either upper control limits or below the lower control limit are marked and considered a signal of changes in the underlying process that are worth further investigation.\n\n# Potential pitfalls\nThe moving ranges involved are [serially correlated](https://bliptext.com/articles/autocorrelation) so runs or cycles can show up on the moving average chart that do not indicate real problems in the underlying process.\nIn some cases, it may be advisable to use the median of the moving range rather than its average, as when the calculated range data contains a few large values that may inflate the estimate of the population's dispersion.\nSome have alleged that departures in normality in the process output significantly reduce the effectiveness of the charts to the point where it may require control limits to be set based on percentiles of the empirically-determined distribution of the process output although this assertion has been consistently refuted.\nSee Footnote 6.\nMany software packages will, given the individuals data, perform all of the needed calculations and plot the results.\nCare should be taken to ensure that the control limits are correctly calculated, per the above and standard texts on SPC.\nIn some cases, the software's default settings may produce incorrect results; in others, user modifications to the settings could result in incorrect results.\nSample data and results are presented by Wheeler for the explicit purpose of testing SPC software.\nPerforming such [software validation](https://bliptext.com/articles/verification-and-validation-software) is generally a good idea with any SPC software.","current_revision":null,"created_by":"kACnuQmH5k9Nr5VNVEpVeg7XpvUmMctD","created_at":"2025-01-26T09:45:08.196Z","updated_at":"2025-01-26T09:45:08.196Z","search_vector":"'chart':4 'control':3 'individu':2 'shewhart':1","search_content":" \n**Name:** Individuals and moving range control chart \n**Proposer:** [Walter A. Shewhart](https://bliptext.com/articles/walter-a-shewhart) \n**Subgroupsize:** n = 1 \n**Measurementtype:** Average quality characteristic per unit \n**Qualitycharacteristictype:** [Variables data](https://bliptext.com/articles/variable-and-attribute-research) \n**Distribution:** [none](https://bliptext.com/articles/indefinite-pronoun) \n**Sizeofshift:** ≥ 1.5σ \n**Varchart:** MR chart for a paired individuals and MR chart.svg \n**Meanchart:** Individuals chart for a paired individuals and MR chart.svg \n**Meanstatistic:** xi \n \n\n# \nIn [statistical quality control](https://bliptext.com/articles/statistical-process-control), the **individual/moving-range chart** is a type of [control chart](https://bliptext.com/articles/control-chart) used to monitor [variables data](https://bliptext.com/articles/variable-and-attribute-research) from a [business](https://bliptext.com/articles/business-process) or [industrial process](https://bliptext.com/articles/list-of-industrial-processes) for which it is impractical to use rational subgroups.\nThe chart is necessary in the following situations:\nThe \"chart\" actually consists of a pair of charts: one, the individuals chart, displays the individual measured values; the other, the moving range chart, displays the difference from one point to the next.\nAs with other control charts, these two charts enable the user to monitor a process for shifts in the process that alter the mean or variance of the measured statistic.\n\n# Interpretation\nAs with other control charts, the individuals and moving range charts consist of points plotted with the control limits, or natural process limits.\nThese limits reflect what the process will deliver without fundamental changes.\nPoints outside of these control limits are signals indicating that the process is not operating as consistently as possible; that some assignable cause has resulted in a change in the process.\nSimilarly, runs of points on one side of the average line should also be interpreted as a signal of some change in the process.\nWhen such signals exist, action should be taken to identify and eliminate them.\nWhen no such signals are present, no changes to the process control variables (i.e. \"tampering\") are necessary or desirable.\n\n# Assumptions\nThe [normal distribution](https://bliptext.com/articles/normal-distribution) is NOT assumed nor required in the calculation of control limits.\nThus making the IndX/mR chart a very robust tool.\nThis is demonstrated by Wheeler using real-world data, and for a number of highly non-normal probability distributions.\n\n# Calculation and plotting\n\n## Calculation of moving range\nThe difference between data point, x_i, and its predecessor, x_{i-1}, is calculated as.\nFor m individual values, there are m-1 ranges.\nNext, the arithmetic mean of these values is calculated as\nIf the data are normally distributed with standard deviation \\sigma then the expected value of is, the [mean absolute difference](https://bliptext.com/articles/mean-absolute-difference) of the normal distribution.\n\n## Calculation of moving range control limit\nThe upper control limit for the range (or upper range limit) is calculated by multiplying the average of the moving range by 3.267:\n.\nThe value 3.267 is taken from the sample size-specific\n D4 \nanti-biasing constant for\n n =2\n, as given in most textbooks on statistical process control (see, for example, Montgomery ).\n\n## Calculation of individuals control limits\nFirst, the average of the individual values is calculated:\n.\nNext, the upper control limit (UCL) and lower control limit (LCL) for the individual values (or upper and lower natural process limits) are calculated by adding or subtracting 2.66 times the average moving range to the process average:\n.\nThe value 2.66 is obtained by dividing 3 by the sample size-specific\n d2 \nanti-biasing constant for\n n =2\n, as given in most textbooks on statistical process control (see, for example, Montgomery ).\n\n## Creation of graphs\nOnce the averages and limits are calculated, all of the individuals data are plotted serially, in the order in which they were recorded.\nTo this plot is added a line at the average value,\n x \nand lines at the\n UCL \nand\n LCL \nvalues.\nOn a separate graph, the calculated ranges\nMRi\nare plotted.\nA line is added for the average value,\n MR \nand second line is plotted for the range upper control limit (\n UCLr \n).\n\n## Analysis\nThe resulting plots are analyzed as for other control charts, using the rules that are deemed appropriate for the process and the desired level of control.\nAt the least, any points above either upper control limits or below the lower control limit are marked and considered a signal of changes in the underlying process that are worth further investigation.\n\n# Potential pitfalls\nThe moving ranges involved are [serially correlated](https://bliptext.com/articles/autocorrelation) so runs or cycles can show up on the moving average chart that do not indicate real problems in the underlying process.\nIn some cases, it may be advisable to use the median of the moving range rather than its average, as when the calculated range data contains a few large values that may inflate the estimate of the population's dispersion.\nSome have alleged that departures in normality in the process output significantly reduce the effectiveness of the charts to the point where it may require control limits to be set based on percentiles of the empirically-determined distribution of the process output although this assertion has been consistently refuted.\nSee Footnote 6.\nMany software packages will, given the individuals data, perform all of the needed calculations and plot the results.\nCare should be taken to ensure that the control limits are correctly calculated, per the above and standard texts on SPC.\nIn some cases, the software's default settings may produce incorrect results; in others, user modifications to the settings could result in incorrect results.\nSample data and results are presented by Wheeler for the explicit purpose of testing SPC software.\nPerforming such [software validation](https://bliptext.com/articles/verification-and-validation-software) is generally a good idea with any SPC software.","revisionCount":0},"hasMore":false,"nextCursor":null}

History