Identify Special Cause Variation

 Identify Special Cause Variation

By OffRoadPilots

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 ofan 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 analysisand 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 problemdefinition, 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 identifyspecial 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, quicklyrespond 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