SPC (Statistical Process Control)


30 Oct 2009

The quality of product is very important to a company. The product or service quality is assured while it's being processed. That's why the process control is an important aspect on producing a product or service.

SPC (Statistical Process Control) is one of process control's tools. By using the SPC as the tool, the process will always be controlled. The tool also helps the company to avoid the deficiency of the products because it will conduct early problem identification.

Statistical Process Control is a process control method which uses data and statistical technique in decision making. The implementation of Statistical Process Control means we use data in every process phase as a basic of decision making or controlling.

Statistical Quality Control (SQC) is statistical technique implementation to control or to solve quality problem (by measuring and improve the quality process). SQC include SPC, Diagnostic Tools, Plan Sampling and the other statistic technique. SPC was needed because there are variations of product or service

Variation is un-uniformity of product or service which is made by the product or service that does not fulfill the specification. Variation or variability is the enemy of quality. The customers always ask for little variation of product and service. That's why; the company should do the improvement and make sure that the variation or variability characteristic of quality of product and service always stay in the line where the customer can tolerate.

The variation caused by a few aspects such as: people, machine, materials, methods, measurement and environment.


  • To minimize the variation which is occurred in the process
  • To improve competition advantages
  • To reduce cost (through the control activity in every process stage)
  • Improving the productivity (by pressing the deficiencies or mistakes)
  • Improving the employee ability to control the process in the field

The SPC Implementation steps in the company are:

  1. Making the process' flowchart
  2. Identify the critical parameter of process
  3. The data collection technique
  4. To implement the statistic control technique (control chart)
  5. To monitor the control process, to review and do action (preventive action)

Flowchart process is a visual chart that shows the process' sequences. The benefit of making flowchart are:

  • To understand the process on creating a product or service
  • To show the complicated part, problem, retarded or unnecessary process, where we can do improvement
  • To compare the reality and the plan
  • To point out the part which needs to collect more detailed data

We need special attention on the process parameter which will influence the product quality or service. The methods to identify the parameter are:

  • Identify the parameter/indicator/factor that affect every steps in the flowchart process
  • choose the critical parameter who affect the product quality which needs special attention or often has a problem
  • The critical parameter should be controlled to watch the obstacle in the process

The parameter example: temperature, pressed, weight, water standard, thickness.

The next step is to collect data from the decided parameter process. There is two types of data. They are:

  1. Attribute Data / Discrete
    The data was collected by counting or conversing into percentage or proportion. Example: The number of complains, number of rejection or waste, damaged proportion or rework proportion
  2. Variable Data / Continuous
    The data is acquired by using the actual measurement on the quality characteristic or the process. Example: long, volume, weight and standard.

After collecting the desired data based on the type of data, the next step is to make control chart.

Control chart is a line chart which uses the control limit as a basis to show whether the process is controlled or not.

The benefits of control chart are:

  • To monitor the variation that is resulted by parameter measuring process
  • To identify early unconformity and take some action before the process is out of control

Control chart has control limit, a line which can show the spread of data and tell the situation in the production, so we can take the right decision.

There are three types of the line control:

  • UCL (Upper Control Limit)
  • LCL (Lower Control Limit)
  • CL (Central Line)

Based on the control limit, we can view the process, whether it is in control or out of control. Process was called "in control" if the point were randomly distributed around the center line and is between the control limit. Meanwhile the process was called out of control, if the points were outside of the control limit.

There are two types control chart in the SPC:

  • Variable Control Chart
  • Attribute Control Chart

Variable control chart is a control chart which uses the variable data on collecting and analyzing the data. There were three methods on creating the control chart, based on the number of the collected data:

  • X bar - R Chart
    The data which were collected in every inspection were grouped into values between 2 and 9. Example: Thickness of solder paste (n=2, 2 hour), Bending strength - kg/cm2 (n=3 per shift), punch terminal pressure (n=5, 2 hour)
  • X bar - S Chart
    The data which were collected in every inspection were grouped into values about 10 or more. Example: The weight of cigarettes, diameter of the cutting result.
  • X Moving Range Chart
    The data is only one value for each inspection. This chart was called moving range because the range was got from the moving data. The data move from one test to another test. This chart is usually used in the twenty four hour industrial section such as cement, chemist manure, oil where the result is relatively homogenous.

The data of this control chart was collected and analyzed by counting. We will find the terms defect / non conformity and defective / nonconforming in this attribute control chart.

The product will be called defect / non conformity if there is at least one specification which is not fulfilled, such as scratch product, waving wall, shells paint.

The product will be called defective/non conforming if the defect causes product rejection and not-functioning, such as broken glass and plates and damaged products.

There are four making control chart methods based on the data collection in the defective inspection such as:

  • P - Chart
    The data was collected and grouped into defect or not and there is difference on the number of subgroup. Example: the chart of number of ATM versus the broken ATM.
  • np - Chart
    The data was collected and grouped into defect or not and the number of subgroup is equal. Example: the landing number of the airplane which did not success every one thousand times of landing.
  • u - chart
    The collected data was the defect of the subgroup. The amount of observed sub group usually is not the same number. Example: the number of present employee versus in discipliner.
  • c - chart
    The collected data was the defect of the subgroup. The amount of observed sub group usually is the same number. The implementation example: Every one hundred telephone lines against the wrong lines.

After defining the control variable, we need to make process capability analysis to predict how consistent the process, to fulfill the specification is. The specification is defined by internal or external customer.

The process was capable if it is able to make one hundred percent output that match with the specification.

Capability index described how far the process fulfilled the firmness specification, it consist of:

  • "Potential" Capability Index (Cp)
  • "Real" Capability Index (Cpk)

Capability Process helps us to focus on the target, wanted by customers. Customer may not be satisfied even though 100% output in the limit specification, and it can cause us to disappeared from the business.

After defining the parameter of the control process, we need to evaluate or audit to the tools, this was done to check if the parameters were suitable. If they are not suitable, we need to prepare some preventive action. (Reference: PQM Consultants, Statistical Process Tools - Public Training, 2005)

Post Comments





Return to previous page


Did You Know?

Malcolm Baldrige National Quality Award is the highest accolode an organization can receive as a demonstration of excellence in quality operations.


Quality is never an accident; it is always the result of intelligent effort

- John Ruskin

"Quality means doing it right when no one is looking"
~ Henry Ford

Quality Is Not An Act, It Is A Habit


If You Want A Quality, Act As If You Already Had It

(Wiliam James)

“Quality means doing it right when no one is looking.”

- Henry Ford -


“Quality is more important than quantity. One home run is much better than two doubles”

- Steve Jobs -

Preference Link