SPC | Statistical Process Control | Quality Core Tool

What is Statistical Process Control (SPC)

SPC is a method of measuring and controlling quality by monitoring the manufacturing process through statistical tools.

In a simple manner, we can say that;

SPC – Controlling the Process with the help of Statistical tools & techniques and hence controlling Quality of Product or Service.

It is one of the 5 quality core tools. (i.e. APQP, PPAP, SPC, MSA & FMEA)

Statistics + Process Control = Statistical Process Control

Comprehensive and detailed information regarding SPC can be obtained from the manual published by the Automotive Industry Action Group (AIAG).

History of Statistical Process Control (SPC)

• In 1924, Walter A. Shewhart at Bell Laboratories developed the                control chart and the concept that a process could be in statistical            control.

• In 1928, first time Statistical Process Control (SPC) Chart was                  introduced in the form of simple graph by Walter A. Shewhart at Bell          Laboratories to monitor and improve the quality of telephones                  manufactured.

• In 1939, he published a book – “Statistical Method from the                      Viewpoint of Quality Control”.

• Widely used during World War II by the military in America at                    weapons facilities.

• Following the World War II, use of SPC techniques in America faded.

• Japanese manufacturing companies adopted the same and is still              using today.

• In the 1950’s, with the effective use of SPC, Deming converted post          war Japan into the world leader of manufacturing excellence.

• Again, during 1970s in America, SPC was popular due to quality                products were imported from Japan.

• Nowadays, SPC is a widely used tool across many industries.

Why Statistical Process Control (SPC)

In the scenario of neck to neck competition, Selling Price plays a vital role in the success of Product or service in the market. To control the Price, we need to optimize our processing costs as Raw Material Cost is not in our Control.

Therefore, focus should be on what is under our control (i.e. Processing Cost). Companies must strive for continuous improvement in quality, efficiency and cost reduction in the processing.

It is better to Prevent then Detect. Here, through SPC, organisation can develop a culture of prevention-based quality control instead of conventional Detection based Quality Control.

By monitoring trends in the process through SPC, we can detect trends or changes in the process which may result in Non-conforming Product.

Benefits of implementing Statistical Process Control (SPC)

It helps in

• monitoring and controlling the Process

• Reducing variation in the Process

• how the process will behave over a period of time (Trends)

• Reducing Rejection and Rework

• Increasing Productivity

• Can assure that it operates at its fullest potential.

• Improving Quality

• Can drive continuous improvement.

Understanding Statistical Process Control (SPC)

For understanding SPC we need to understand “Statistics” & “Process Control”

Statistics:

It is a branch of mathematics concerned with collecting, organizing and interpreting Data.

Product or process measurement data (readings) is collected.

Collected data is used to evaluate, monitor and control a process.

As statistics is related to Data, let us understand Types of Data

Type of Data

Data collected can be of Two Types:

1. Variable Data or Continuous Data

2. Attribute Data or Discrete Data

Variable Data or Continuous Data

We get Measurement output on a continuous scale. (can be measured)

Examples:

> 25.40mm Diameter of Rod,

> 28.54 kg of Metal weight,

> 30.45 sec. time for Processing a product, etc..

Attribute Data or Discrete Data

We get inspection output as decision or counting.

Examples:

> OK or Not OK,

> Good or Bad,

> Pass or Fail,

> No. of Defective Parts,

> No. of defects on painted part, etc..

What is Process Control?

Process:

It is a set of interrelated or interacting activities which transforms inputs into outputs.

Examples:

> Cutting,

> Turning,

> Drilling,

> Painting, etc..

Process Control:

Evaluation of Variation of a Process over a period of time. It monitors stability of a process.

Example:

In turning process, Shaft Diameter variation was within ±0.2mm on 01.02.2019 which gradually increased to ±0.3mm on 20.02.2019.

Variation:

Law of Nature: No two objects or things are exactly alike.

This law is also applicable to parts produced in manufacturing. If we measure it with higher precision instruments, no two parts will be truly identical. There will be a difference present. This difference is termed as Variation.

Sources of Variation:

Understanding Variation

Causes of Variation

Common Cause

Variation in the process which is always present.

•    Process is stable and repeatable when only Common cause is present.

•    Needs management actions to correct.

•    85% of total causes are common causes.

Special Cause

Variation in the process caused by External or unusual events.

•   It makes the process unpredictable.
•   Needs local actions to correct
•   Almost 15% of total causes are special causes

Measurement of Variation

If we want to reduce the Variation then we need to measure the variation and current status of Variation present in Data

•   Measurement of Central Tendency

•   Measurement of Dispersion

Measurement of Central Tendency of Data

It is the measurement of data set which tends towards a central value. i.e. finding central value of Data set to which it tends to.

FIGURE

Measures of Central Tendency are

1.   Mean (Average)

2.   Median

3.   Mode

Mean (Average) is the most commonly used Measure of Central Tendency

Measurement of Dispersion of Data

The extent of the spread of the values from the mean value.

Measures of Dispersion are

1.   Range (R)

2.   Standard Deviation (s)

3.   Variance (S2)

4.   Co-efficient of Variation (CV)

Standard Deviation (s) is the most commonly used Measurement of Dispersion

Statistical Tools used for Process Control

Below mentioned are the tools used for Process Control.

(Also known as 7 QC Tools)

• Check sheet

• Histogram

• Pareto chart

• Flow chart

• Cause and Effect Diagram

• Control chart

• Scatter Diagram

Control chart is an effective tool to monitor the process

Control Charts and its types

For Variable Data

• Xbar – R chart

• Xbar – MR chart

• Xbar – S chart

For Attribute Data

• p chart

• np chart

• c chart

• u chart

How typical Control Chart Looks like?

How to select Control Chart?

Thank You