Skip to main content

Statistical process control and related quality management process

  • Author:
  • Updated date:

In recent years, quality management has been given priority in the action agenda of many companies. Before beginning to implement an inclusive Total Quality Management, the company needs to examine many different factors including the process variation, process capability, different controlling and monitoring techniques, the organization culture and different international standards to guide its quality management improvement process. Statistical process control is one of the most effective ways to supervise and control the process and ultimately the product quality via a statistical approach.


An overview of the statistical process control (SPC)

Statistical process control is one of the tools that an organization can use to determine whether a process works the way it is intended to do. In any process it is inevitable to have some degree of variation. Variation in the production process can cause defects in the product, reduce the consistency of product quality and can negatively influence “quality” of the product or service. SPC allows manager to analyse whether the variation is random and natural, or it is caused by some assignable causes and needs to be corrected. SPC monitors and controls the process to ensure that it operates at its full potential.

SPC typically consists of two steps. The first step is to randomly examine a sample of the output from a process. The second step is to make the strategic decision as to whether the process is producing products with “quality” conforming to a set of prearranged values. The section aims to propose a framework to successfully implement the SPC procedure into an organization, explain various characteristics of the model and how it overcomes the challenges posed by the process variation.

Factors affecting the implementation of SPC

In literature, organizational and social factors are most frequently attributed as the most common reasons for the failure of the implementation of SPC. Some common reasons for the failure to implement the SPC process thoroughly include the lack of commitment in the management and operation of the process, lack of training to help employees understand the SPC techniques, and the declining attention after the first introduction of SPC. Furthermore, some researchers point out that the execution of the SPC process can be very challenging because it requires a great amount of time and money to implement SPC company-wide, a constant demand for support from top management, the dedicated involvement of the whole organization from the top to the lowest levels, the need for a comprehensive database to determine the relevant quality concept and issue, the in-depth knowledge of managers and experts about the process, and other teamwork and project management skills.

Proposed process to successfully implement the SPC into an organization

In order to successfully implement the SPC into an organization, this is a process that an organisation can consider implementing:

• Phase 1: Plan. To begin with, the organization should spend an adequate amount of time to identify the quality problem of the process, possible causes, and possible solutions to eliminate the problems. This phase helps to give the organization a general idea of whether it needs to implement the SPC process, and accordingly plans the amount of time and money needed for the process.

• Phase 2: Awareness. In this phase, the organization should start to educate its employees as well as managers about the necessity of implementing the process and gain their approvals of participating thoroughly in the process.

• Phase 3: Experimental sample. In this phase, the organization implements its plan on a small scale, preferably with some selected focus groups. Each group will follow a specific activity agenda including 4 steps: 1) Identifying the problems and weak points, 2) Search for ways to improve the weakness, 3) Define effective measurements and control methods to control the process, 4) Carry out the process and evaluate the results. After this phase it will be decided whether the process is appropriate and can be continued or not. If the process is decided that it is not satisfying, changes will be made, and phase 3 will be repeated until producing a satisfying outcome.

• Phase 4: Increase the number of experimental samples to cover other key processes of the organization as well. In this phase, other groups of different departments/processes are also invited to participate in the activities of phase 3 of the process. The organization will learn how to coordinate different groups and bring all the people together.

• Phase 5: Study the impacts of the process. At this point, the organization will conduct a comprehensive analysis of the whole process to ascertain that it helps improve the quality problems without causing any other problems. This will be the benchmark to decide whether the process can be applied on a larger scale for the whole organization.

• Phase 6: Implement the process in the whole organization. The process will be used for the whole organization and will be evaluated on a regular basis its effectiveness in solving quality problems.

In the proposed process, phase 3 remains one of the most critical phase determining the success of the process. The biggest challenge facing in this phase is to find an effective method to detect control the process variation, which lead to the reduction in the process variation. There are two causes of variation. The first cause is called common, or random, cause of variation which happens naturally and unavoidably during the process, and we cannot identify or control it. The second type of variation is called assignable causes of variation. These causes can be identified, corrected or avoided if an organization implements an appropriate solutions or quality management system.

In quality control, it is necessary to distinguish between the random variations and the assignable variations. For example, we conduct an experiment and find out that the average length of a tape is 10 inches. We determine that the amount of natural variation of tape length is between 9.8 and 10.2 inches. Accordingly, we would control the production process to ensure the length of tapes we produce stays within this range. If production goes out of this range—tapes are found to have a length of 10.5 inches — this would indicate that there is a problem with the process because the variation is greater than the natural random variation.

Model statistical process control implementation process

Model statistical process control implementation process

Using control charts to monitor process variation

One analytical tool managers can use to monitor the process variation is control chart. By analysing the control chart, managers can decide if the process variation is caused by the common causes or some special assignable causes of variation. A control chart usually takes the form of a graph that shows whether a sample of data falls within the common or normal range of variation. A control chart has upper and lower control limits. The value that lies under the lower limit or above the upper limits indicates an assignable variation. In practice, the formula to calculate the upper and lower control limits on a control chart is the mean ± 3 standard deviations.

The characteristics that control charts measure can be classified into two groups: variables and attributes. A control chart for variables is used when the characteristics can be measured and represented by a real number such as the thickness, age, and speed. On the other hand, a control chart for attributes is used when the characteristics take on discrete values such as origin, number of people, or type of materials. However, regardless of the type of characteristics, in order to construct a control chart, we can follow the same steps including process definition, data collection, chart design, data plot, chart evaluation, and process modification. Depending on the goal of the process, type of product and characteristics of the process, the type of chart and specific design of chart can be varied.

Examples of control charts

Examples of control charts

Scroll to Continue

Control charts

Steps to construct a control chart:

  • Process definition
  • Data collection
  • Chart design
  • Data plot
  • Chart evaluation
  • Process modification.

Understanding process capability

Another important factor of SPC is the process capability, which evaluates the ability of a production process to satisfactorily meet a predetermined quality. Process capability can be computed by the process capability index, which provides a mechanism to analyse and forecast process performance. Process capability can also be presented in the form of graphs such as histograms and plots to visualize the distribution of the data and decide if the process is working properly. Since capability analysis highlights changes and tendencies of the system during production, analysing process capability helps to determine the ability for manufacturing between tolerance limits and engineering specifications.

Utilizing the benchmarking schemes

Understanding an organization’s process capability is the key for the leaders to set the benchmark achievements for their organizations. Benchmarking is the process in which an organisation compares its base data regarding its costs and performance against the data from some other organizations. The benchmarking process can be used externally by comparing an organisation with another organisation or the average of a group of companies or used internally by comparing one department in a company with another department, or comparing the branch of a company in one country/city with its branch in another country/city.

Depending on the goals and missions of an organization, it can set different benchmark performance standards. Following are some example criteria that can be used as benchmark performance standards:

• Timeliness: The benchmark sets the goal that the company should provide predetermined types of information within a specified timeframe. Timeliness is important to benchmark users, sponsors and stakeholders.

• Efficiency: The organisation sets the goal to improve its efficiency to match with the higher level of efficiency of another organisation/ department.

• Defect containment: The process should minimize number of defects, and should not introduce any new errors into the process. Analysis of benchmark data should consider that there exists some unavoidable variation in the data that occurs naturally due to some differences in the process.

• Process conformance: The execution of process activities should conform to any plans and procedures developed to describe the intended process. This may be judged by quality management audits or process capability assessments.


Nowadays offering high quality product gives an organization a competitive advantage ahead of its competitor. Therefore, controlling and improving quality has become an important business strategy for many organizations.

SPC can be used successfully in various kinds of organizations, ranging from mass-production of simple products, small batch production of various medium complex products, to low volume production of complex products. The methodological part of the implementation of the process is subject to tailoring owing to differences in the situation. Although the goals of the process presented are quite universal, some ways to tailor the method to suit the situation should be addressed. The SPC concept eventually can also be applied to other processes in the organization, thus paving the way for the Total Quality Management.

Statistical process control: Honda - a success story


HH (author) from Vietnam on January 29, 2015:

Hi, yes I did. I have a Master's Degree in Economics and I did take several management courses over years.

peachy from Home Sweet Home on January 28, 2015:

are you studying management related course?

Related Articles