Skip to main content
Advantive

Where do the typical control chart signals come from?

By Grace Barton Updated
Where do the typical control chart signals come from?

The purpose of control charting is to regularly monitor a process so that significant process changes may be detected.

These process changes may be a shift in the process average (Xbar) or a change in the amount of variation in the process. The variation observed when the process is operating normally is called common cause variation. When a process change occurs, then special cause variation occurs.

Control charting captures snapshots of the process average and variation over time. By first establishing the variation we expect from the process (via control limits) when it is stable (in control), we are able to detect subsequent process changes. When specific signals are observed on the control charts, we conclude that the process is unstable (change occurred, out of control) because the probability of observing those signals if the process had not changed is very small.

Hypothesis Testing

In a statistical hypothesis test, we presume some statement, which is called the null hypothesis (H0). We also establish the alternative hypothesis (H1), which is what we are actually attempting to conclude (if the data supports it). Every time a new value is plotted on a control chart, a hypothesis is evaluated. The initial assumption is that the process is stable (in control). If, after plotting a point, we have enough evidence to reject this null hypothesis (we see a signal), we conclude that the alternate hypothesis (the process is out of control) is true.

Ideally, we would correctly reject the H0 every time the process is actually out of control. However, if the process is actually out of control, and we do not detect it, we have made a Type II error. This may be a severe error since the process has changed but we haven’t noticed it and don’t react. An appropriate sample size may be selected to minimize the occurrence of a Type II error (see the previous “Ask the Expert” article on determining sample sizes for Xbar charts).

We would also ideally not reject the H0, if in fact the process has not changed. However, if we observe a signal even though the process has not changed, we have made a Type I error (α). This error leads to inefficiency since we will react to a signal but not find any actual cause, the process having not actually changed. By convention, the probability of a Type I error (α) is specified as 0.0027 (0.27%). This results in the control limits trapping 99.73% of the statistic that is being plotted on the control chart.

Note: 99.73% equates to ±3 standard deviations from the process average, if the data being plotted is normally distributed.

Basis of Chart Signals

Now it should be clear that Hypothesis Testing is performed in order to determine whether sufficient evidence exists to conclude that the process is unstable. The common rules for identifying process instability are based on the probability of observing such signals assuming the process is actually stable.

For example, suppose we observe a single point that falls outside the upper control limit on an Xbar chart as shown below:

The probability that we would observe a sample average that is more than 3 standard deviations away from the process average assuming the process is stable is only 0.0027 (since the control limits trap 99.73% of the sample averages). Because this probability is so small, we conclude that the alternate hypothesis is true and we react as though the process is unstable. Of course, we might be wrong in which case a Type I error has occurred.

Another common chart signal is a run of 7 (or more) consecutive points above or below the center line. This is illustrated below.

So, what is the probability of actually observing 7 points in a row above (or below) the center line assuming that the process is actually stable?

For any random sample, the probability of getting a sample average above the process average is simply ½ . There is also a 50% chance of seeing a sample average below the process average.

To obtain 7 points in a row above the centerline, we need to find the probability of the following: one point being above the centerline AND the next point being above the centerline AND the next point being above the centerline and so on until 7 is reached. Essentially, it’s like flipping a coin and getting 7 consecutive heads. Since these events are independent (if the process is stable), the joint probability of getting 7 consecutive heads is simply ½ x ½ x ½ x ½ x ½ x ½ x ½ = 1/128 = 0.0078.

Thus, the probability of seeing this pattern is very small (less than 1% chance) if the process stable. Therefore, we reject this null hypothesis and conclude that the process is unstable. (In this case, it seems that the process average has shifted).

Many other rules are used by practitioners to detect trends or process shifts on Xbar charts. They include:

  • 7 points in a row trending upward or downward
  • 14 points alternating up and down
  • 2 out of 3 consecutive points more than 2 standard deviations away from the centerline on the same side of the chart
  • 4 out of 5 consecutive points more than 1 standard deviation away from the centerline on the same side of the chart
  • 15 consecutive points within +/- 1 standard deviation of the centerline

The probabilities of observing many these patterns assuming a stable process are not extremely difficult to compute, but it suffices to say that the probabilities are low. Since it’s unlikely the pattern comes from a stable process, we conclude that the process is unstable and react to it.

Note that some people use slightly different rules. For example, waiting for a run of 8 points above the centerline (rather than 7). This version of the rule will result is fewer Type I errors but a greater number of Type II errors. All rules should balance the errors that can be made in interpreting control charts.

Steven Wachs, Principal Statistician
Integral Concepts, Inc.

Integral Concepts provides consulting services and training in the application of quantitative methods to understand, predict, and optimize product designs, manufacturing operations, and product reliability. www.integral-concepts.com

Grace Barton

Marketing Specialist

About the Author Latest Posts

Grace Barton is a digital marketing and competitive intelligence professional who crafts strategic narratives by bridging marketing insights with analytical expertise. At Advantive, she creates engaging, data-driven content tailored to the distribution, manufacturing, packaging, and quality industries. Her goal is to deliver impactful messaging that drives engagement and growth based on specific gap closure needs, whether responding to sales organization requirements, pinpointing gaps in content, or meeting immediate market trends.
She thrives on transforming competitive intelligence into actionable insights for the sales organization. Grace manages Advantive’s competitive intelligence platform, Klue, to equip the sales team with the battlecards and market data they need to stay ahead of competitors. Since launch, she’s built 28+ battlecards across four lines of business, ensuring the GTM strategy stays sharp.
Grace has a passion for leveraging market insights with storytelling to guide strategic decision-making, empower sales organizations, and nurture organizational growth.

Areas of Expertise: Digital Marketing, Competitive Intelligence, Strategic Narratives, Marketing Insights, Analytical Expertise

LinkedIn

Editorial standards

Fact-checking & editorial guidelines

Every article on advantive.com is written and reviewed against an internal accuracy standard before publication. Here's what that means in practice.

  • Product claims are verified by the brand team that owns the platform.

    When an article references InfinityQS, WinSPC, PQ Systems, Pinpoint, ParityFactory, ProPlanner, KiwiPlan, DDI System, VeraCore, or any of Advantive's other specialty platforms, the relevant product team checks technical statements about features, deployment, and current capability before the article goes live.

  • External statistics cite their source inline.

    When an article references industry survey results, regulatory benchmarks, or third-party research, the source is linked at the point of citation. Statements without an inline source link are first-party observations drawn from Advantive's product teams or customer base.

  • Publication and revision dates stay visible.

    The original publication date and the most recent revision date are both shown on every article. Topics that change quickly — AI capabilities, regulatory rules, product roadmaps — are revisited on a tighter cadence than evergreen reference content.

  • Corrections are issued openly.

    If a factual error is reported, the article is updated, the revision date advances, and material corrections are noted at the bottom of the article so readers can see what changed and when.

Found something wrong, or have a citation to add? Get in touch with the editorial team and we'll review it.

Subject-matter review

Reviewed by subject-matter experts

Advantive is a portfolio of 14+ specialty software platforms — each one built and maintained by a product team that has spent years inside a specific manufacturing or distribution discipline. Articles in technical channels are reviewed by the relevant team before publication.

  • Quality & SPC content

    Reviewed by the InfinityQS, WinSPC, and PQ Systems product teams — the platforms behind statistical process control, capability analysis, and gage management deployments across food, automotive, pharma, and CPG manufacturers.

  • Manufacturing operations & MES content

    Reviewed by the PINpoint, ProPlanner, ParityFactory, and VIA Information Tools teams, whose platforms run production scheduling, traceability, and shop-floor execution for discrete, automotive, and food-and-beverage manufacturers.

  • Packaging & converter content

    Reviewed by the KiwiPlan, Abaca, and AdvantZware teams, who build software specifically for corrugated, folding-carton, and packaging-converter operations.

  • Distribution, ERP & B2B commerce content

    Reviewed by the DDI System (inFORM ERP), Distribution One, VeraCore, Pepperi, and Commerce Vision teams, whose platforms run wholesale ERP, fulfillment, field sales, and B2B portals for specialty distributors and 3PLs.

Are you a practitioner with domain expertise to contribute? Get in touch — we accept guest contributions from operators in the industries we serve.