Md Atikur Rahman

Control Charts
A Control Chart is a graphical representation used to study how a process changes over time. It plots
data points in the time order and helps detect trends or shifts in the process by comparing them to the
statistically calculated control limits.

Components of a Control Chart

• Center Line (CL): This is the average or mean of the data points being plotted. It represents the
target or expected value for the process.
• Upper Control Limit (UCL): This is the maximum value a data point can reach before it indicates
that the process might be going out of control. It is usually set at 3 standard deviations above the
center line.
• Lower Control Limit (LCL): This is the minimum value a data point can reach before signaling
potential issues with the process. It is typically 3 standard deviations below the center line.

How Control Charts Work?

The Control charts plot data points over time with center line representing the process average and
upper and lower control limits acting as thresholds for the variation expected within the process.

• If all points lie within the control limits and there is no discernible pattern the process is
considered in the control.
• If a point falls outside the control limits or there is a non-random pattern it signals that the
process may be out of the control and corrective actions should be taken.

A control chart helps differentiate between the common cause variation and special cause variation.

Types of Control Charts

The Control charts are broadly categorized into two types based on the nature of the data:

• Variable Control Charts

• Attribute Control Charts

Variable Control Charts

These charts are used when the data being monitored is continuous. The most common variable
control charts include:

• X-bar Chart: Monitors the mean of a process over time.

• R-bar chart: Monitors the range of the process.
• S-bar chart: Tracks the standard deviation of the process.

Attribute Control Charts

The Attribute control charts are used for the discrete data such as the counts of defects or defective
units. The most common attribute control charts are:

• P-bar chart: Tracks the proportion of defective items in the sample.

• NP-chart: Monitors the number of defective items in the fixed-size sample.
• C-bar chart: Tracks the count of defects in the unit.
 Md Atikur Rahman

• U-chart: Monitors the number of defects per unit in the sample of the varying sizes.

Construction of a Control Chart

To create a control chart, follow these steps:

Step 1: Data Collection

First, collect data for the process we're monitoring. This data is typically collected over a set period of
the time and must be representative of the process in the question.

Step 2: Calculate Process Average

The central line (CL) of the control chart represents the process average in which is calculated by taking
the mean of the collected data points.

Step 3: Determine Control Limits

The Control limits are calculated to identify the boundaries for the acceptable variation. These limits
are typically set at three standard deviations above and below the process average. The Upper Control
Limit (UCL) and Lower Control Limit (LCL) can be calculated as:

UCL = CL + 3σ and LCL = CL − 3σ

where σ is the standard deviation of the data points.

Step 4: Plot the Data

Once the central line and control limits are established the data points are plotted over time. This visual
representation helps in the determining if the process is in the control or if corrective action is required.

How to Interpret Control Charts

A control chart provides valuable insights into the process's behavior. The key indicators to monitor are:

• In Control Process: If the data points are fluctuating within control limits without forming any
specific pattern the process is considered stable or in control.
• Out of Control Process: If the data points fall outside the control limits or form recognizable
patterns it indicates that the process may be experiencing special causes of the variation
requiring further investigation.

Applications of Control Charts

The Control charts have wide applications in the various industries:

• Manufacturing: Used to monitor product quality ensuring that the dimensions or specifications
of the products remain consistent over time.
• Healthcare: The Control charts are used to track patient outcomes or medical treatment
efficacy allowing for the early detection of the problems in healthcare processes.
• IT and Software Development: They help monitor system performance metrics like response
time or error rates enabling the identification of the trends that might affect system reliability.
• Financial Services: The Control charts can track key financial metrics over time such as the
transaction volumes or fraud detection rates to the ensure the processes are under control.
 Md Atikur Rahman

Solved Questions on Control Charts

Question 1: A factory produces 100 units daily and following samples are taken over 5 days: [10,
12, 11, 9, 13]. Calculate the control limits for the X-bar chart.

Solution:

Calculate the mean:

10+12+11+9+13
Mean = = 11
5

Calculate the standard deviation:

(10−11)2 +(12−11)2 +(11−11)2 +(9−11)2 +(13−11)2

σ=� = 1.414
4

Control limits:

UCL = 11+3×1.414= 15.24

LCL = 11-3×1.414= 6.76
Therefore, the UCL is 15.24 and LCL is 6.76.

Question 2: A sample of 50 products is taken every day for 7 days. The number of defective
products recorded over 7 days is: [3, 2, 1, 4, 3, 5, 2]. Construct a P-chart.

Solution:

Proportion of defective products per sample:

3 2
𝑝𝑝1 = = 0.06, 𝑝𝑝2 = = 0.04, ….
50 50

Calculate the average proportion −𝐩𝐩 / P bar

0.06+0.04+0.02+0.08+0.06+0.10+0.04
−
p = = 0.057
7
 Md Atikur Rahman

Control limits:
−(1−−)
0.057(1−0.057)
UCL = +3× �
p p
−
p = 0.057 +3× � = 0.138
𝑛𝑛 50

−(1−−)
0.057(1−0.057)
LCL = −p -3× �
p p
= 0.057 -3× � = 0 (adjusted)
𝑛𝑛 50

Thus, the UCL is 0.138 and LCL is 0.

Question 3: A process has the following ranges of the 5 samples: [2.5, 3.1, 2.9, 3.3, 3.0]. Calculate the
control limits for the R-chart.

Solution:

Calculate the average range (𝑅𝑅_ )

_ 𝟐𝟐.𝟓𝟓+𝟑𝟑.𝟏𝟏+𝟐𝟐.𝟗𝟗+𝟑𝟑.𝟑𝟑+𝟑𝟑.𝟎𝟎
𝑹𝑹
= = 2.96
𝟓𝟓

Control limits for R-chart (using factors for sample size (n=5), (D3 = 0), (D4 = 2.114):

UCL=𝑫𝑫𝟒𝟒 × 𝑹𝑹_ = 2.114×2.96 = 6.26

LCL=𝑫𝑫𝟑𝟑 × 𝑹𝑹_ = 0×2.96 = 0

So, the UCL is 6.26 and LCL is 0.

 Md Atikur Rahman

Question 4: A process monitors the number of defects per day in the production unit and counts for 5
days are: [4, 6, 3, 5, 7]. Construct a C-chart.

Solution:
_
Calculate the average number of defects (𝒄𝒄):

_ 𝟒𝟒+𝟔𝟔+𝟑𝟑+𝟓𝟓+𝟕𝟕
𝐜𝐜
= =5
𝟓𝟓

Control limits for the C-chart:

UCL= 𝐜𝐜_ + 𝟑𝟑�𝐜𝐜_ = 5+𝟑𝟑√𝟓𝟓 =11.71

_ 𝟑𝟑 _
LCL= -
𝐜𝐜 �𝐜𝐜
= 5 -𝟑𝟑√𝟓𝟓 = 0 (adjusted)

Thus, the UCL is 11.71 and LCL is 0.