InsightFab
Knowledge Base/I-MR Chart: Still Able to Create Control Charts Even with Only One Data Point Per Batch
SPC7 min read

I-MR Chart: Still Able to Create Control Charts Even with Only One Data Point Per Batch

When only one sample can be collected per production batch, using an X-bar R Chart in Minitab results in an error due to insufficient subgroup size. The I-MR Chart is the appropriate tool for such scenarios, as it does not require multiple samples per subgroup.

Scenario

You are responsible for monitoring a chemical batch process where one barrel is produced per batch, and one sample is taken to measure viscosity. Your boss says: "Help me create a control chart." You open Minitab, select X-bar R Chart, input the data, and the software reports an error: "Subgroup size must be ≥ 2." You don't know which one to choose.

Plain Language Explanation

The X-bar R Chart requires multiple measurements per subgroup (typically n = 4 or 5). However, in many situations, there is only one data point per time period: daily yield, viscosity per batch, monthly energy consumption.

In this situation, use the I-MR Chart (Individual - Moving Range Chart).

  • I Chart: Monitors each individual measurement, setting UCL/CL/LCL
  • MR Chart: Monitors "the difference between two adjacent points"—i.e., the short-term variation

The control limits for I-MR use the average of the moving range to estimate process variation, eliminating the need for subgroups.

Practical Judgment

Scenarios suitable for I-MR:

  • Only one measurement per day/shift (daily yield, daily output)
  • Destructive testing (only one sample can be taken per batch)
  • Chemical batch processes (one quality indicator per batch)
  • Equipment maintenance cycle records

Scenarios not suitable for I-MR:

  • Opportunity to take multiple samples within a short period → Switch to X-bar R
  • Highly autocorrelated data (today's value strongly predicts tomorrow's) → Requires time series methods

Limitations of I-MR: Higher normality requirement than X-bar R; if data is severely skewed, control limits will be inaccurate, so a normality test should be performed first.

How InsightFab Works

Upload single-column data, and the system automatically detects "this is single measurement data," suggesting switching to an I-MR Chart, so you don't need to remember that this function exists.

Golden Quote

"Not having multiple samples doesn't mean you can't create a control chart; it means you're unaware of the I-MR Chart's existence."

Want to try it yourself?

Every tool mentioned in this article is available on InsightFab — just upload a CSV to analyze.

Go to Tools

More in this category

The Red Dot on the Control Chart Is Not Bad News—It's Your First Clue to Finding the Problem

When a red dot appears on an X-bar R control chart, is your initial response to exclude it and continue production? Such an action could lead to three hours of additional defective products. The red dot on a control chart signifies the process communicating for the first time.

Measurement System Requirements for Control Charts: MSA Before SPC

In process control, abnormal data on control charts often originates from measurement system variation. This article emphasizes that Measurement System Analysis (MSA) should precede the implementation of Statistical Process Control (SPC) to ensure the accuracy and reliability of measurement data. MSA, by evaluating key items such as repeatability, reproducibility, bias, linearity, and stability, prevents misinterpreting process abnormalities and wasting resources.

CUSUM and EWMA Control Charts: Powerful Tools for Detecting Small Shifts

Traditional Shewhart control charts quickly detect large shifts, but often miss subtle process drifts. CUSUM and EWMA are specifically designed to monitor and detect these small shifts; this article explains their application and interpretation.