In-Depth Analysis of the Seven QC Tools · Histogram

By: QTank Published: 5/3/2026 Views: 1017

Current rating: ★★★☆☆ Rate this Equivalent to 9 ratings (from visitors: 1)

Introduction

Relying solely on the average can deceive you twice.

Two work teams have an average nonconformity rate of 3%, but one team's rate is consistently between 2.5% and 3.5%, while the other fluctuates dramatically from 0.5% to 8%—which one can you accept?

The average only reveals one aspect of the data. A histogram, however, is the mirror that uncovers the true distribution of the data.

Chapter 1: The Essence of Histograms

1.1 What is a Histogram?

A histogram is a statistical tool that displays the distribution of data using bars. It divides the range of data values into several equal-width intervals (bins) and counts the frequency of data points in each interval, represented by the height of the bars.

Core Logic:
  Continuous data → Group by intervals → Count frequency of each group → Draw bars

  By observing the shape of the bars, you can determine:
  Where is the center of the data? (mean/median)
  How dispersed is the data? (range/standard deviation)
  What is the distribution pattern? (normal? skewed? bimodal?)
  Does the data exceed specification limits? (is it within tolerance?)

1.2 Why the Average Isn't Enough

Assume two sets of data, both with an average of 50:

  Data Set A: 49, 50, 51, 49, 50, 51, 49, 50, 51
    → Average = 50, with minimal variation

  Data Set B: 10, 30, 50, 70, 90, 40, 60, 20, 80
    → Average = 50, but with significant variation

Both data sets have the same average, yet their quality levels are vastly different!

The average is just one characteristic of the data. A histogram can tell you much more:

Where is the center of the data? (mean/median)
How dispersed is the data? (range/standard deviation)
What is the distribution pattern? (normal? skewed? bimodal?)
Does the data exceed specification limits? (is it within tolerance?)

1.3 Three Major Functions of Histograms

Function	Description	Applicable Scenarios
Understanding Process Status	Determine if the process is stable and normal	Process capability analysis, initial process validation
Identifying Abnormal Patterns	Detect abnormal patterns such as bimodal, skewed, or isolated peaks	Process anomaly diagnosis
Evaluating Process Capability	Compare with specification limits to calculate CP/CPK	SPC, process capability studies

1.4 Histogram vs. Bar Chart

Many people confuse histograms with bar charts, but they are different:

Dimension	Histogram	Bar Chart
Data Type	Continuous data (length, weight, temperature)	Categorical data (nonconforming types, work teams)
X-Axis	Numerical intervals (continuous)	Category names (discrete)
Bar Arrangement	Bars are arranged in numerical order, cannot be rearranged	Bars can be arranged in any order
Bar Spacing	No spacing (continuous distribution)	Spacing (categorical)
Content Displayed	Distribution pattern	Comparison of magnitudes

Chapter 2: Drawing and Interpreting Histograms

2.1 Seven Steps to Drawing a Histogram

Step 1: Collect Data
  Collect at least 50 data points, preferably more than 100

Step 2: Determine Data Range
  Maximum value - Minimum value = Range (R)

Step 3: Determine Number of Groups
  Empirical formula: Number of groups k = √n or k = 1 + 3.3lg(n)
  Common reference: 50 data points → 7 groups, 100 → 10 groups, 200 → 12 groups

Step 4: Determine Group Width
  Group width = Range / Number of groups (round to a convenient number for calculation)

Step 5: Determine Group Boundaries
  Upper and lower limits of each group, without overlap

Step 6: Count Frequencies
  Count the number of data points in each interval

Step 7: Draw the Histogram
  X-axis = Numerical intervals, Y-axis = Frequency
  Draw the bars and mark the specification limits (if applicable)

2.2 Typical Shapes of Histograms

① Normal (Bell-shaped)
Frequency ^
 7 |          █          
 6 |          █          
 5 |       █  █  █       
 4 |       █  █  █       
 3 |    █  █  █  █  █    
 2 | █  █  █  █  █  █  █ 
 1 | █  █  █  █  █  █  █ 
   +—————————————————————→ Numerical Intervals
  Process is normal, most common

② Skewed (Left-skewed/Right-skewed)

  Right-skewed (Skewed to the left)
Frequency ^
 6 | █                   
 5 | █  █                
 4 | █  █  █             
 3 | █  █  █  █          
 2 | █  █  █  █  █       
 1 | █  █  █  █  █  █  █ 
   +—————————————————————→ Numerical Intervals

  Left-skewed (Skewed to the right)
Frequency ^
 6 |                   █ 
 5 |                █  █ 
 4 |             █  █  █ 
 3 |          █  █  █  █ 
 2 |       █  █  █  █  █ 
 1 | █  █  █  █  █  █  █ 
   +—————————————————————→ Numerical Intervals

③ Bimodal
Frequency ^
 6 |    █           █    
 5 |    █           █    
 4 |    █           █    
 3 |    █  █     █  █    
 2 | █  █  █  █  █  █  █ 
 1 | █  █  █  █  █  █  █ 
   +—————————————————————→ Numerical Intervals
  Two different distributions mixed together
  Cause: Different equipment/shifts/materials

④ Flat-top
Frequency ^
 5 | █  █  █  █  █  █  █ 
 4 | █  █  █  █  █  █  █ 
 3 | █  █  █  █  █  █  █ 
 2 | █  █  █  █  █  █  █ 
 1 | █  █  █  █  █  █  █ 
   +—————————————————————→ Numerical Intervals
  Multiple data sets with different means mixed

⑤ Isolated Peak
Frequency ^
 7 |          █                
 6 |       █  █  █             
 5 |       █  █  █             
 4 |    █  █  █  █  █        █ 
 3 |    █  █  █  █  █        █ 
 2 | █  █  █  █  █  █  █     █ 
 1 | █  █  █  █  █  █  █  █  █ 
   +———————————————————————————→ Numerical Intervals
  A few data points deviate from the main body
  Cause: Temporary anomaly, measurement error

⑥ Jagged
Frequency ^
 6 |             █             
 5 |       █     █     █       
 4 | █     █     █     █     █ 
 3 | █     █  █  █  █  █     █ 
 2 | █  █  █  █  █  █  █  █  █ 
 1 | █  █  █  █  █  █  █  █  █ 
   +———————————————————————————→ Numerical Intervals
  Improper group width setting or rounding issues

Chapter 3: Practical Cases of Histograms

Case 1: Manufacturing — Distribution of Shaft Diameter

Background: A machining workshop measures the diameter of shafts, with a specification requirement of φ50±0.1mm
100 data points were collected

Histogram Analysis:
  → Normal distribution, center at 50.02mm (slightly above the upper limit)
  → All data within the specification limits (49.90-50.10mm)
  → Distribution width is about 60% of the specification width

Conclusion:
  ✓ Process capability is sufficient (CPK≈1.2)
  ✓ However, the center is slightly above the upper limit, monitor tool wear trend

Suggestion:
  Set the target value to 50.00mm, allowing for tool wear

Case 2: Bimodal — Mixed Data from Different Equipment

Background: A plastic injection molding workshop collected product weight data
The histogram shows a clear bimodal distribution

Analysis:
  After stratifying by equipment:
  → Equipment A: Average weight 102g
  → Equipment B: Average weight 98g
  → Significant difference in average weights between the two pieces of equipment

Root Cause:
  Different temperature settings on the two pieces of equipment

Action:
  Standardize the process parameters of the two pieces of equipment
  After adjustment, the histogram returns to a normal distribution

Case 3: Skewed — Service Industry Waiting Time

Background: Data on customer waiting times at a bank counter

The histogram shows a clear right-skewed distribution (long tail on the right)

Analysis:
  → Most customers wait between 3-8 minutes
  → But a few customers wait more than 20 minutes
  → This is a typical "long-tail distribution" in the service industry

Action:
  ── Set a 15-minute over-warning
  ── Increase counters during peak hours
  ── Conduct a special analysis for customers with excessively long waiting times

Case 4: Isolated Peak — Identifying Measurement Errors

Background: An electronics factory measures resistance values

The main body of the histogram is normally distributed between 100-105Ω
But there is an isolated peak at 95Ω (3 data points)

Investigation Revealed:
  → These 3 data points came from a new inspector
  → The inspector used an uncalibrated measuring tool

Action:
  ── Recalibrate the measuring tool
  ── Retrain the inspector
  ── Recall and re-measure the data from this batch

Chapter 4: Common Misconceptions About Histograms

Misconception 1: Insufficient Sample Size

× Incorrect Practice:
  Only collected 20 data points to draw a histogram
  → Bars are uneven, unable to see the true distribution

✓ Correct Practice:
  Collect at least 50 data points, preferably more than 100
  The more samples, the clearer the distribution pattern

Misconception 2: Improper Number of Groups

× Too Few Groups:
  Only divided into 3-4 groups, losing a lot of information
  Unable to see distribution details

× Too Many Groups:
  Divided into more than 20 groups, too many bars, distribution is chaotic
  Unable to see the overall trend

✓ Correct Practice:
  Determine the number of groups using the formula k = √n or empirical methods
  50 data points → 7 groups, 100 → 10 groups

Misconception 3: Not Comparing with Specification Limits

× Incorrect Practice:
  Only draw the histogram, without marking the upper and lower specification limits
  → Even if you see the distribution, you don't know if it's within tolerance

✓ Correct Practice:
  Mark the upper and lower specification limits (USL/LSL) on the histogram
  Intuitively judge whether the process capability is sufficient

Misconception 4: Not Investigating Abnormal Patterns

× Incorrect Practice:
  See bimodal, skewed, or isolated peaks and just leave them
  → Miss the opportunity to identify issues

✓ Correct Practice:
  Abnormal patterns are clues!
  Bimodal → Mixed data from different sources → Stratified analysis
  Isolated peaks → Special causes → Investigate root causes
  Skewed → Unstable process → Parameter adjustment

Misconception 5: Focusing Only on Shape, Not on Data Volume

× Incorrect Practice:
  30 data points, draw a beautiful "normal distribution"
  → Assume the process is good

✓ Correct Practice:
  Small samples may accidentally present any pattern
  The larger the sample size, the more reliable the conclusion
  First, use a control chart to determine if the process is in control

Chapter 5: Combining Histograms with Other Tools

5.1 Histogram + Control Chart

Best Partner:

  Control Chart → Determine if the process is in control (time dimension)
  Histogram → Determine the distribution pattern and position (static dimension)

Combined Use:
  ① First, use a control chart to determine if the process is in control
  ② After the process is in control, use a histogram to analyze the distribution pattern
  ③ Compare with specification limits, evaluate process capability
  ④ Continuously monitor with a control chart

5.2 Histogram + Stratification

Combined Use:

  Step 1: Draw an overall histogram to identify abnormal patterns (e.g., bimodal)
  Step 2: Use stratification to categorize by equipment/shifts/materials
  Step 3: Draw a histogram for each category
  Step 4: Identify the source of differences and implement targeted improvements

  → The histogram identifies "issues," stratification finds "where"

5.3 Histogram + Normality Test

Two ways to determine if data follows a normal distribution:

  Qualitative: Check if the histogram is bell-shaped
  Quantitative: Perform a normality test (e.g., Anderson-Darling test)

Why is it important:
  → If the data does not follow a normal distribution
  → The formulas for calculating CPK and CP may not be applicable
  → Data transformation or non-parametric methods may be needed first

5.4 Histogram + Specification Limits → Process Capability Index

Prerequisites for calculating the process capability index (CP/CPK):
  ① The process is in control (confirmed by control chart)
  ② The data follows a normal distribution (confirmed by histogram)

CP = (USL - LSL) / 6σ
CPK = min(CPL, CPU)
  where CPL = (μ - LSL) / 3σ
       CPU = (USL - μ) / 3σ

Judgment Criteria:
  CPK ≥ 1.33 → Sufficient process capability
  1.0 ≤ CPK < 1.33 → Adequate process capability, but requires attention
  CPK < 1.0 → Insufficient process capability, requires improvement

Chapter 6: Advanced Uses of Histograms

6.1 Histogram + Specification Limits (Process Capability Visualization)

Best Practice Histogram Includes:
  ── Bars: Data distribution
  ── Two vertical lines: USL (upper specification limit) and LSL (lower specification limit)
  ── One vertical line: Target value
  ── Annotations: Mean, standard deviation, CPK value

  → One chart can determine if the process meets requirements

6.2 Before and After Improvement Comparison Histograms

Pre-Improvement Histogram:
  ── Distribution center deviates from the target
  ── Wide distribution
  ── Data points exceed specification limits

Post-Improvement Histogram:
  ── Distribution center aligns with the target
  ── Narrower distribution
  ── All data points within specification limits

  → Two charts side by side, intuitively showing the improvement effect

6.3 Grouped Histograms

Display histograms for different groups on the same chart using different colors
(Use semi-transparent fills to see overlapping areas)

Value:
  ── Intuitively compare distribution differences between groups
  ── Quickly identify which group needs improvement
  ── Avoid misjudgment of bimodal distributions

Chapter 7: Evaluation Standards for Histograms

Evaluation Dimension	Good Standard	Poor Performance
Adequate Sample Size	More than 100 data points	Less than 50
Reasonable Number of Groups	Number of groups determined by formula	Too many or too few groups
Clear Scale	Clear and labeled axes	Confused scale, no units
Specification Marking	Mark USL/LSL/target value	No specification limits
Correct Analysis	Combine shape, position, and dispersion	Focus only on shape, ignore position
Action Output	Investigate and address abnormal patterns	End after drawing

Summary: The "Way" and "Method" of Histograms

Method (How to Draw):
  ── Collect more than 100 data points
  ── Determine the number of groups and group width
  ── Draw bars and mark specification limits
  ── Observe the distribution pattern

Way (Why to Draw):
  ── Not just to create a "pretty distribution chart"
  ── But to "see the truth behind the average"
  ── To "identify issues in the data and find improvement directions in the distribution"

The greatest value of a histogram is not in "describing data," but in "unveiling the lies of the average."

A manager who focuses only on the average may be overseeing a highly unstable process—while believing everything is fine.

The histogram is the tool that helps him see the truth.

Document Version: v1.0 Generated Date: 2026-05-03 Author: Excellence Quality Think Tank

Issue 6: Histogram (Histogram)