Defining Statistical Process Control
Statistical process control (SPC) entails applying statistical techniques to measure, monitor, and manage production methods or processes. It uses 14 statistical and analytical tools (7-QC and 7-SUPP) to control process inputs (independent variables).
7-QC refers to seven quality control tools developed by Dr Kaoru Ishikawa in 1974 while creating a text Guide to Quality Control. They include the cause-and-effect/Ishikawa/fishbone diagram, control chart, scatter diagram, check sheet, Pareto chart, histogram, and stratification (Suman & Prajapati, 2018). 7-SUPP refers to the seven supplemental tools used to enhance process improvement further. They include defect maps, data stratification, process flowcharts, event logs, randomization, progress centers, and sample size determination.
The history of quality control dates back to the late 13th century. The journey was started by artisans in medieval Europe who formed quality unions called guilds. The craftsmanship model was carried into the 19th century and evolved through the mid-1750s to the factory system in Great Britain, grounded on product inspection.
The Industrial Revolution took over in the early 1800s under the Taylor system in America through enhanced production methods. World War II enhanced the need for quality improvement to increase productivity while guaranteeing the safety of military products. The U.S. Armed Forces adopted sampling inspection instead of unit-by-unit inspection.
Walter Shewhart started the focus on quality in processes rather than on the final product in the 1920s. W. Edwards Deming and Joseph M. Juran continued to enhance quality improvement in Japan after WWII, leading to total quality management (TQM). The U.S. later introduced vital quality initiatives, including the ISO 9000 series of quality-management standards, the Malcolm Baldrige National Quality Award, and the Baldrige National Quality Program. Organizations, including Motorola, contributed to quality improvement by creating additional SPC techniques, such as Six Sigma.
Control Charts
The center line represents the mean value of observations of a process or production recorded over time. The center line’s deviation shows process variation, measured by the standard deviation of observations (Young & Smith, 2022). In the example above, the center line is the mean, while the deviation is represented by the space above and below the line.
The control limits of a control chart represent variation that helps to show whether a process is stable and predictable. If a process is not predictable, the limits help to show when the process or production is out of control. Control limits are marked by two horizontal lines, one above and the other below the center line, representing the upper and lower limits, respectively.
The upper and lower control limits are based on random variations in the process or production. They are typically established by multiplying the standard deviation by three (Young & Smith, 2022). However, an organization can use customer requirements to define the specification limits. A process can be in control, but it may not be able to meet the specifications.
Common Cause Variation
Processes or production within an organization are likely to have variations in the system. Common causes are variations ingrained in a system over time, such that they are expected. They affect everyone working on the design and the system’s results, showing a stable statistical control process.
Since organizations constantly seek to improve the quality of their process and products, the typical cause case can be handled by determining whether it is good enough. The quality improvement team seeks ways to narrow the gap between the upper and lower control limits (Slyngstad, 2021). The proposed interventions aim to transform the system to produce different results. New control limits have to be established for quality control and involve the application of rapid-cycle PDSA.
Special causes are the opposite of common causes, representing unusual occurrences in a process or production. Given that they are not expected, special causes occur due to unique circumstances. They show that the system is unstable and there is a lack of statistical control. Improvement in the production process is mandatory. Positive special causes should be replicated to become part of the system, while the negative causes should be eliminated to avoid recurrence and bring the system into statistical control (Young & Smith, 2022). The subject matter experts must be involved in identifying the reasons for special causes.
Special Cause Variation
Special causes of variation can be signaled by observing several distinctive patterns appearing on the control chart. One of the most common patterns is a single point spotted beyond the control limits (Gupta, 2021). This may be caused by a sudden, significant shift from the mean that occurs only once.
A second primary signal for special cause variation is a continuous trend in one direction, either upwards or downwards. Oakland and Oakland (2019) state that the trend cannot be easily defined since the pattern largely depends on the type of process, such as manufacturing or service processes. Another common special cause variation is represented by small shifts that are maintained over time around the center line. The more minor shifts could result from a change in raw material or work instruction, differences in shift, or measurement device/calibration (Gupta, 2021). Other factors include gaining skills at work, changing the setup procedure, or the maintenance program.
SPC in a Business Context
SPC is mainly applied in manufacturing to support process improvement. It helps to define if a process is under control from a statistical point of view, making it possible to detect production problems. In this way, only a tiny part of the production process can be stopped to rectify the situation instead of closing the whole system.
Regarding better outcomes, SPC helps guarantee the uniformity of the final product by reducing scraps and defects. Manufacturing entities can apply control charts to display statistical information to monitor quality in a graphic format (Kear, 2020). The graph can show unusual process variations and help in management decision-making to establish control over baseline standards, improve the manufacturing process, and identify and correct flaws based on factual information.
Service organizations have increasingly adopted SPC to reap its benefits. SPC can enhance effectiveness by analyzing process behavior and identifying hidden problems. This helps indicate required actions for continuous improvement and predict whether processes can attain better results by meeting the objectives (Brito et al., 2018). The measures used in SPC in service companies can generate data that can be analyzed to obtain pertinent information about what needs to be done. The information can support decision-making across various processes, such as quality assurance and testing.
Process Variation
Common cause variation can contribute to decision-making by applying interventions to improve processes to attain better performance. If the process improvement initiatives are maintained, the organization can achieve stabilization at a higher level of performance (Brito et al., 2018). Stabilized critical processes are a feature of a high-maturity organization or one that aims to achieve the highest maturity level.
Special cause variation can incite the management to investigate the cause and seek options to reduce it by monitoring processes and outcomes. Control charts provide real-time information from which the management can learn and manage variation over time. The administration can use the data to compare past, present, and desired performance, seek to reduce undesired variation, and push for the desired variation (Slyngstad, 2021). Additionally, the management can implement best practices and use SPC to benchmark their performance.
Benefits and Drawbacks
Statistical Process Control and, in particular, control charts can be applied in the healthcare business to improve a hospital’s emergency department that wants to monitor and improve the waiting time for patients. Using the collected data, the hospital creates a control chart for waiting times. They may use statistical tools such as an X-bar chart or an individual/moving range (I/MR) chart. The control chart will have a central line representing the target waiting time or mean, as well as upper and lower control limits based on acceptable variation.
By utilizing SPC in a healthcare setting, hospitals can proactively manage waiting times, identify process inefficiencies, and make data-driven improvements. This can lead to enhanced patient experiences, reduced waiting times, and improved overall operational efficiency in delivering healthcare services. SPC supports data-driven decision-making through the collection and analysis of data over time (Oakland & Oakland, 2019). It provides objective and reliable information about a process’s performance, allowing decision-makers to base their decisions on data rather than assumptions or subjective opinions. This data-driven approach promotes more informed and accurate decision-making.
It facilitates early detection of process deviations through monitoring process performance and detecting deviations from the desired standards or specifications. By plotting data on control charts and setting control limits, SPC enables decision-makers to identify and address process variations early on. Early detection allows for timely intervention and corrective actions to prevent potential quality issues or service failures.
One major drawback of SPC is that the application of control charts in signaling the need for process improvement does not guarantee its success. The control chart does not connect with clinical control as it only indicates the absence of special cause variation. SPC also carries many risks as it may not be faultless and cannot solve all problems, mainly when not applied wisely.
In some instances, it may lead to undesirable outcomes in cases where the standard decision rules to detect special causes are not correctly used to identify process changes (Young & Smith, 2022). The worst-case scenario involves arriving at wrong conclusions about process performance, which amounts to wasting time, effort, and spirit. When medical errors are committed, this can contribute to patient harm.
References
Brito, D. F., Barcellos, M. P., & Santos, G. (2018). Investigating measures for applying statistical process control in software organizations. Journal of Software Engineering Research and Development, 6(1). Web.
Gupta, B. C. (2021). Statistical Quality Control. John Wiley & Sons.
Kear, F. W. (2020). Statistical process control in manufacturing practice. CRC Press.
NSW Government – Clinical Health Commission. (n.d.). Control charts. Web.
Oakland, J. S., & Oakland, R. J. (2019). Statistical process control. Routledge.
Suman, G., & Prajapati, D. (2018). Control chart applications in healthcare: A literature review. International Journal of Metrology and Quality Engineering, 9, 5. Web.
Slyngstad, L. (2021). The contribution of variable control charts to quality improvement in healthcare: A literature review. Journal of Healthcare Leadership, Volume 13, 221–230. Web.
Young, M., & Smith, M. (2022). Standards and evaluation of healthcare quality, safety, and person-centered care. StatPearls. Web.