Identify Special Cause Variation

Special cause variation, also known as assignable cause variation, refers to variation in a process that can be traced to specific, identifiable causes or factors. These causes are not part of the normal, expected variability that occurs in a process. Identifying special cause variation is a critical aspect of process improvement and quality control. It involves distinguishing between random or common cause variation and variation that can be attributed to specific factors. 

Key characteristics and methods for identifying special cause variation are control charts, unusual data points, patterns in data, data clustering, data outside specification limits, outliers, sudden changes in inputs, observations from process operators, root cause analysis, and historical data comparison. 

Statistical control charts are graphical tools used in statistical process control to monitor process performance over time. They typically have upper and lower control limits that represent the acceptable range of variation for a process. If data points on a control chart fall outside these control limits, it suggests the presence of special cause variation. The control chart is a graph used to study how a process changes over time. Data are plotted in time order. A control chart always has a central line for the average, an upper line for the upper control limit, and a lower line for the lower control limit. These lines are determined from historical data. By comparing current data to these lines conclusions can be drawn about whether the process variation is consistent (in control) or is unpredictable (out of control) An out of control process is affected by special causes of variation. Statistical Process Control by using control charts, and analysis are considered one of the seven basic quality control tools. These seven tools are Cause-and-effect diagram, Check sheet, Control chart, Histogram, Pareto chart, Scatter diagram, and Stratification chart.

Control charts for variable data are used in pairs. The top chart monitors the average, or the centering of the distribution of data from the process. Control charts for attribute data are used singly. 

Unusual data points are data points that fall outside the control limits on a control chart. Control limits are typically set at ±3 standard deviations (6 Sigma) from the process mean. Data points beyond these limits suggest special cause variation.

Patterns in data is to examine data for patterns or trends that are not part of the normal process behavior. Common patterns to watch for include spikes, shifts, cycles, and abrupt changes. 

Data clustering is when data points cluster together in a non-random way. This is an indication of the presence of a special cause variation. For example, if a machine consistently produces parts with dimensions clustered around a specific value, this could be a special cause. If reports, or data received from independent sources, but are identical, is an indication of a special cause variation. Raw data does not come is clusters and when it happens is an indicator of a corrupted process. When an SMS enterprise operates with corrupted processes their root cause analysis and corrective action plans have failed before they are implemented. 

A corrupt process refers to a situation where a system or a specific procedure within an organization has been compromised or tainted by unethical, illegal, or dishonest activities. Corruption in a process can have severe consequences, including financial losses, damage to reputation, legal repercussions, and a breakdown in trust. A corrupt process does not paint the true picture of the system but does however paint a true picture of an SMS enterprise. An example of a corrupt process is the checkbox syndrome, or when the primary task becomes to complete all checkboxes as opposed to the discovery of special cause variations. Another example of a corrupt process is when third-party SMS programs maintain control of an SMS enterprise’s SMS. Third-party process compliance as a substitute to an operator’s SMS processes is recognized by allowability to operate with a flexible and suitable SMS.

Data points outside specification limits are data points that consistently fall outside the defined specification limits for a product, service, or process. This indicates the presence of a special cause variation and suggests that the process is out of control and that it is not meeting its intended requirements.

An outlier is an observation or data point that significantly deviates from the rest of the data in a dataset. In other words, it's a data point that is unusually distant or different from the majority of the data points in a sample. Outliers can occur for various reasons, including measurement errors, data entry errors, natural variability, or genuinely exceptional cases. An outlier is a special cause variation. The SMS enterprise’s role and responsibility by an accountable executive is to identify and investigate outliers in the data. 

Sudden changes in process inputs, such as airport daily inspections, aircraft pre and post flight inspections, or sudden change in processes are special cause variations. When there are sudden changes in these inputs, initiate an investigation of their impact on the process and perform a root cause analysis.  

Observations from process operators are inputs and reports from personnel who are conducting the tasks. They are the frontline worker who have firsthand knowledge of the process. This knowledge is invaluable data for an SMS enterprise to run with a successful SMS. Experienced operators may notice unusual events or changes that could lead to special cause variations.

Perform a thorough root cause analysis to identify the specific factors or events responsible for the variation. Tools like the 5-Whys and Fishbone (Ishikawa) diagrams, can help uncover underlying causes.

A root cause analysis (RCA) is a systematic process for identifying the underlying causes of a problem or an event, with the goal of addressing the root causes rather than just the symptoms. It is a valuable problem-solving technique used in the airline industry, both airports and airlines, safety management system quality control, project management, e.g. plan of construction operations. The primary objective of RCA is to prevent the recurrence of problems by addressing their fundamental causes. However, after implementation of a corrective action plan, a new special cause variation may appear. 

A root cause analysis may involve multiple rounds of analysis and corrective action. It encourages a proactive approach to problem-solving and continuous improvement within an SMS enterprise. By addressing the root causes of problems, rather than just symptoms, airports and airlines can prevent issues from recurring and improve their overall performance and reliability.

Applying the appropriate steps in an acceptable sequence is a criteria for a successful root cause analysis.

1. The first step is to define the problem or event. Clearly and precisely define the problem or event that needs to be analyzed. Write down in details what the problem or event is. This step involves gathering information, data, and evidence related to the issue.

2. The second step is to gather data and information. Collect relevant data and information related to the problem or event. This may involve reviewing records, conducting interviews, and using various data collection methods.

3. The third step is to identify immediate causes. Determine the immediate or proximate causes, or causal factors of the problem or event. These are the factors that directly contributed to the issue and are often the most apparent.

4. The fourth step is to identify contributing factors. An SMS enterprise needs to look beyond the immediate causes to identify the factors that contributed to the problem. These factors may include human factors, organizational factors, supervision factors or environmental factors, equipment failures, or external influences.

5. The fifth step is to construct a cause-and-effect diagram (Fishbone Diagram) or apply the 5-Why process. A cause-and-effect diagram, also known as a fishbone diagram or Ishikawa diagram, is a graphical tool used to visualize the possible causes of a problem. It helps organize and categorize factors into groups, such as people, processes, equipment, materials, and environment. The 5-Why root cause process is to ask a Why-question five times, or more, to determine the root cause. The very first answer to a Why-question opens only one of many doors, becomes the cornerstone of the root cause analysis and determines the final answer, no matter how many times it’s asked. A simple 5-Why root cause analysis should consider asking the question How? A How-question is unbiased, neutral to the event and provide data to be collected for quality control and eventually quality assurance. It is crucial for a successful 5-Why root cause analysis that the matrix is a 5x5 Why-matrix.   

6. The sixth step is to identify the root cause. Among the contributing factors identified, pinpoint the root cause, or the fundamental factor or systemic issues that, if addressed, would prevent the problem from recurring. A root cause is often hidden beneath the surface and require deeper analysis.

7. The seventh step is to validate the root cause. Verify the identified root cause using data and evidence. Ensure that the root cause is responsible for the problem. 

8. The eighth step is to develop and implement a corrective action. In order to preserve the integrity of a post-CAP analysis, only one corrective action is implemented to determine its effect on the process. Once the root causes are confirmed, develop a corrective action or solution to address the root cause. This action should be practical, feasible, and aimed at preventing future occurrences of the problem.

9. The ninth step is to monitor and follow-up. Implement the corrective action and monitor the effectiveness over time. Monitoring is to monitor the process each time it is put into action. E.g. if a CAP was implemented to an airport’s daily inspection, monitoring would be initiated at the onset of the daily inspection process. Follow up to ensure that the same problem does not recur and that the solutions are sustainable. Be aware that a new special cause variation may cause a process deviation, which cannot be assigned to the current CAP.

10. The tenth step is documentation. Document the entire root cause analysis process, including the problem definition, data collected, causes identified, corrective actions taken, and outcomes. This documentation is invaluable for tracking progress and lessons learned.

Historical data comparison is to compare current data to historical data to determine if the observed variation is unique or has occurred before. If it's unique, it's more likely to be a special cause variation. Historical data should go back at least seven years to include the triennial audit requirements data retention. 

Identifying and addressing special cause variation is crucial for process improvement and maintaining product quality. Once special causes are identified, corrective actions can be taken to eliminate or mitigate their effects, leading to more stable and predictable processes. Control charts and statistical process control (SPC) methods are often used to monitor and identify special cause variations in industrial and manufacturing processes, service processes, and aviation safety performance processes. 

If a special cause variation is insignificant to operations and an immediate CAP fixed the problem, leave the process untouched and monitor for repetition. Overcontrolling a process to make it perfect deliver a less desired process outcome than an imperfect process. Use safety critical areas and safety critical functions to determine the severity of impact or processes. 

An example of an insignificant special cause variation is a flat tire upon landing. The outcome could be a runway incursion, but a flat tire is insignificant to the approach and landing process since it may only occur at a ratio of 0.07% of all movements. Just as a nail on the highway causing one car tire to go flat is insignificant to the highway travel process. An insignificant event becomes significant to the process when it produces a trend identified in an SPC control charts. When a trend is identified, a special cause variation becomes significant, and a root cause analysis is required.    

It is my industry experience that control charts are uncommon tools for SMS enterprises and other aviation related industries. Different types of charts are used, but the analyses of these charts are based on emotions. An operator may only accept column charts or pie charts where the only criteria is to reduce the number of events. One operator who was dissatisfied with the number of events increased the trigger level and immediately the number of events were reduced. Control charts are impartial, they are unbiased, they are anchored to a data platform only, and emotions and opinions are eliminated from the equation.  

Special cause variations are neutral and does not care about the variation itself.  

Using SPC control charts, such as spcforexcel.com, and integrate the SiteDocs.com model is a winning team for a successful safety management system. When applying statistical process control an SMS enterprise, such as an airport and airline, established an analysis process with reliability and integrity. When using SPC control chart, the process does not change with a change of personnel responsible for SMS oversight, e.g. accountable executive, a change of person responsible for operational quality control, e.g. SMS manager, or trigger a change in the quality assurance process. When using approach charts the identification process remain in control and is stable. (Which is what we want.)  

It is possible to use other methods than mathematical designed SPC control charts to identify special cause variations. When applying the manual method, the responsible person must fully comprehend the process, all common cause variations, and its interaction with other processes every step of the way from input to outcome. When using the manual method to identify special cause variations, physically observe the process or system to identify any unusual behaviors or conditions causing the special cause variation.

Physically observing a process is process tracking. Process tracking i also known as process monitoring or process control, refers to the practice of continuously monitoring and managing various aspects of a process to ensure it operates effectively, efficiently, and within predefined parameters. This systematic approach helps SMS enterprises to maintain consistency, improve quality, and identify and address deviations or issues in real-time. Process tracking can be applied to any airport or airline processes. tries, including manufacturing, healthcare, finance, and project management. 

Process tracking includes data collection. To track a process, data must be collected from various sources within the process. Data collected may encompass performance metrics, key indicators, and other relevant information.

Manual process tracking includes real-time monitoring. Process tracking involves monitoring the ongoing performance of the process in real-time or near-real-time. This allows for immediate awareness of any anomalies or deviations from the desired performance standards.

Manual process tracking identifies key performance indicators (KPIs). Identify and define KPIs that represent safety critical areas and safety critical functions. These KPIs serve as benchmarks and can be used to measure the process's effectiveness and efficiency.

Manual process tracking includes thresholds and alarms. Establish predetermined thresholds or alarm limits for KPIs. When data points cross these limits, it triggers alerts or alarms, notifying responsible personnel or systems of potential issues.

Manual process tracking includes a manual analysis of collected data to identify patterns, trends, and potential areas for improvement. 

Manual process tracking includes process optimization. Based on data analysis, process tracking can lead to continuous process improvement initiatives. These improvements can enhance efficiency and reduce costs.

Manual process tracking includes documentation and to maintain detailed records of the tracked process data, including any actions taken in response to deviations or alarms. Proper documentation is essential for analysis, reporting, and compliance purposes.

Manual process tracking includes visualization, visual representations of process data, such as charts, graphs, and dashboards, can provide a clear and intuitive overview of process performance, making it easier to spot issues.

Manual process tracking includes regulatory compliance assessments. The aviation industry, both airports and airlines include strict regulatory requirements, and process tracking is crucial for ensuring compliance with standards and regulations.

Manual process tracking includes feedback and reporting. Regularly communicate process performance results and improvements to relevant stakeholders, including management, airline and airport personnel, and customers, as appropriate.

Process tracking plays a pivotal role in quality control, operational efficiency, and risk management. It helps SMS enterprises to maintain control over their processes, quickly respond to deviations, and make data-driven decisions to continually improve their operations.

Special cause variations are identified in a system analysis prior to implementation of a new process, or a new system. A pre-analysis takes into account the probability of a special cause variation to occur, and identification of probable variations to occur. Based on probable variations (multiple variations needs to be analysed) perform a root-cause analysis and a risk analysis. The report is then submitted to the accountable executive to accept or reject. A post-analysis of a special cause variation is to perform a root cause analysis, including a risk analysis of the actual deviation, or drift. The report with recommendations is then submitted to the accountable executive for review to accept or reject the recommendation.    

OffRoadPilots