2.75.5 Control Charts

More information:


Control charts creates a graph used to determine if a process is being controlled under stable conditions. Origin provides different kinds of controls based on what data are collected

Processing

Select from the drop-down list and click on the icon to open dialog:

CC GUI.png

Variables Charts for Subgroups

The control charts in this list are to monitor the process performance of continuous data, which are collected from subgroups. Six (6) types of control charts are provided:

  • Xbar-R
  • Xbar-S
  • I-MR-R/S
  • Xbar
  • R
  • S

Please look at the tables below for when and how they should be used:

Combination Charts
Charts Components Purpose Usage Comments
Xbar-R An Xbar and an R chart Determine if your process is stable For subgroups that have 2~8 observations Examine the R chart first to make sure the process variation is in control; otherwise, the result of the Xbar chart may be inaccurate
Xbar-S An Xbar and an S chart Determine if your process is stable For subgroups that have 9 or more observations Examine the S chart first to make sure the process variation is in control; otherwise, the result of the Xbar chart may be inaccurate
I-MR-R/S An individuals chart, a moving range chart, and an R or S chart. Assess the variation within and across subgroups Consistent source of variation within the subgroups The moving range charts shows whether the between-subgroup variation is in control, and the R or S chart shows whether within-subgroup variation is in control
Basic Charts
Charts Description
Xbar Plots the process mean over time for variables data in subgroups
R Plots the process range over time for variables data in subgroups
S Plots the process standard deviation over time for variables data in subgroups

Variables Chart for Individuals

The control charts in this list are to monitor the process performance of continuous data but not in subgroups. Four (4) types of control charts are provided:

  • I-MR
  • Z-MR
  • Individuals
  • Moving Range

Please look at the table below for when and how they should be used

Combination Charts
Charts Components Usage
I-MR An individual chart and a Moving Range chart For continuous data that are individual observations not in subgroups
Z-MR A standardized process mean chart and a Moving Range chart For short-run processes, where little data are available
Basic Charts
Charts Components Usage
Individuals Also called I-Chart. Plots individual observations over time for variables data. Monitor the process center for individual observations
Moving Range Also called MR-Chart. Plots the moving range over time for individual observations. Monitor the process variation

Attributes Charts

The control charts in this list are to monitor the process performance of attribute data, such as the counts of defectives for defects. Eight (8) types of charts are provided:

  • P Chart Diagnostic
  • P
  • Laney P'
  • NP
  • U Chart Diagnostic
  • U
  • Laney U'
  • C

Please look at the table below for when and how they should be used:

know only whether each item is defective or not
Use when Charts Description
Subgroup size is unequal P Chart Diagnostic Test for overdispersion or underdispersion in your defects data
P Plots the fraction, percent, or proportion of nonconforming units
Laney P' Variation of P Chart, can adjust for overdispersion or underdispersion in your data
Subgroup size is equal NP Plots the number of nonconforming units when subgroup sizes are equal.
know the number of defects on each item
Use when Charts Description
Subgroup size is unequal U Chart Diagnostic Test for overdispersion or underdispersion in your defects data
U Plots the number of nonconformities or defects per unit
Laney U' Variation of U Chart, can adjust for overdispersion or underdispersion in your data
Subgroup size is equal C Plots the number of nonconformities or defects when subgroup sizes are equal.

Time-Weighted Charts

The Time-Weighted Charts are useful in detecting small process shifts. Three (3) types of charts are provided:

  • Moving Averag
  • EWMA
  • CUSUM

Please look at the table below for when and how they should be used

Chart Description Comments
Moving Average Also called MA chart. Plots the unweighted moving average over time for individual observations Useful in detecting small process shifts.
EWMA Plots the exponentially weighted moving averages. Each EWMA point incorporates information from all the previous subgroups or observations based on a user-defined weighting factor. Generally preferred over MA chart as all previous observations are taken into account.
CUSUM Displays the cumulative sums (CUSUMs) of the deviations of each sample value from the target value Useful in detecting processes that are slowly slipping away from the target value due to machine wear, calibration issues, etc.


Multivariate Charts

Multivariate control charts track several related process variables at once to spot abnormal behavior. They're the multivariate version of Xbar-R or I-MR charts, essential when variables are connected. Four (4) types of charts are provided:

  • T²-Generalized Variance
  • T²
  • Generalized Variance
  • Multivariate EWMA

Please look at the table below for when and how they should be used

Combination Charts
Chart Description Usage
T²-Generalized Variance Mean + Variability Use to detect if your process is stable
Multivariate EWMA Mean vector (smoothed) Sensitive for the small mean shifts
Basic Charts
Chart Description
T² Monitors only the process mean vector (location) using Hotelling's T² statistic.
Generalized Variance Monitors only the process variability (covariance matrix Σ) using the determinant of the sample covariance matrix
English | Deutsch | 日本語