Control Plan and Sustenance

Learning Objectives

By the end of this lesson, the learner will be able to:

Explain the purpose and importance of a Control Plan.
Describe the relationship between a Control Plan and the DMAIC methodology.
Identify the essential elements of an effective Control Plan.
Define process characteristics, CTQs, specifications, and control methods.
Establish appropriate measurement and monitoring methods.
Define sampling frequency and sample size.
Establish reaction plans for out-of-control or nonconforming conditions.
Distinguish preventive controls from detection controls.
Connect Control Plans with SPC, standard work, FMEA, and process documentation.
Explain how Control Plans support process ownership and sustained improvement.
Develop a practical Control Plan for a process.
Monitor process performance and update controls when process conditions change.

1. Introduction

Improvement is not complete simply because a process has achieved better results.

A Six Sigma project must ensure that the improvement continues after the project team reduces its involvement.

This is the purpose of process control and sustenance.

A process may initially improve because:

The Green Belt is actively monitoring it.
The project team is reviewing performance frequently.
Operators have received additional attention.
Temporary corrective actions are being maintained.
Management is closely involved.

However, once the project is formally closed, the process must continue to perform without depending on the project team.

A Control Plan provides a structured method for documenting how important process characteristics will be controlled and monitored.

It answers questions such as:

What must be controlled?
What is the required standard?
How will it be measured?
How frequently will it be measured?
Who is responsible?
What happens when the process goes outside the defined condition?
How will the response be documented?

The Control Plan therefore converts improvement results into a repeatable operating discipline.

2. Concept / Theory

2.1 What Is a Control Plan?

A Control Plan is a documented plan that identifies the important characteristics of a process and specifies how those characteristics will be controlled, measured, monitored, and responded to.

A Control Plan normally connects:

Process Step → Characteristic → Requirement → Measurement → Control Method → Sampling → Reaction Plan → Responsibility

The exact format may differ between organizations, industries, and standards, but the underlying purpose remains the same.

2.2 Why Control Plans Are Important

Without a Control Plan, an improvement can gradually deteriorate.

For example:

A company improves customer-service response time from 24 hours to 8 hours.

The project is declared successful.

Six months later:

Staffing changes.
The workload increases.
The response-time monitoring is discontinued.
New employees are not trained in the improved process.
Escalation rules are no longer followed.

Response time gradually returns to 18 hours.

The original improvement has not been sustained.

A Control Plan helps prevent this by defining how the improved process will continue to be monitored and managed.

2.3 Control Plans and DMAIC

The Control Plan is closely associated with the Control phase of DMAIC.

However, control thinking should begin earlier in the project.

Define

The project identifies:

Customer requirements
CTQs
Business requirements
Process objectives

Measure

The team establishes:

Operational definitions
Measurement methods
Baseline performance
Data-collection procedures

Analyze

The team identifies:

Root causes
Critical process inputs
Sources of variation

Improve

The team:

Implements solutions.
Validates improvement.
Establishes improved process conditions.

Control

The team:

Standardizes the improved process.
Establishes monitoring methods.
Develops reaction plans.
Transfers ownership to the process owner.
Monitors performance.

The Control Plan therefore acts as an important bridge between improvement and sustained operational performance.

2.4 Process Control vs Product Inspection

An effective Control Plan should not depend only on final inspection.

There are two broad approaches:

Detection

Detection identifies a problem after it occurs.

Examples:

Final inspection
End-of-line testing
Review of completed transactions
Customer complaint monitoring

Prevention

Prevention reduces the likelihood that the problem will occur.

Examples:

Error-proofing
Automated validation
Standard work
Interlocks
Preventive maintenance
Training
Approved operating parameters

A strong process-control system uses prevention wherever practical and uses detection as an additional safeguard.

2.5 CTQs and Critical Process Characteristics

A Control Plan should focus attention on characteristics that matter.

These may include:

Critical-to-Quality (CTQ) characteristics
Critical process parameters
Customer requirements
Regulatory requirements
Safety characteristics
Key performance indicators
Characteristics linked to significant failure modes

Not every process variable needs the same level of control.

Control effort should be proportionate to:

Customer impact
Risk
Process sensitivity
Failure consequences
Historical performance
Regulatory requirements

2.6 Control Methods

A Control Plan can specify different types of controls.

Examples include:

Standard operating procedures
Checklists
Automated alarms
Control charts
Visual controls
Inspection
Sampling
Error-proofing
Preventive maintenance
Training certification
Approval requirements
Automated system validation

The control method should be appropriate to the characteristic being controlled.

3. Key Topics

3.1 Essential Elements of a Control Plan

A practical Control Plan may include the following fields:

Element	Purpose
Process Step	Identifies where control occurs
Process Input / Output	Defines what enters or leaves the step
Characteristic	Identifies what must be controlled
CTQ / Criticality	Identifies importance
Specification / Standard	Defines acceptable requirement
Measurement Method	Defines how performance is measured
Measurement Device / System	Identifies equipment or method
Sample Size	Defines how many observations are required
Sampling Frequency	Defines how often measurement occurs
Control Method	Defines how performance is controlled
Control Chart / Statistical Method	Defines analytical monitoring where applicable
Responsible Person	Assigns ownership
Reaction Plan	Defines response to abnormal conditions
Records	Defines documentation requirements

Organizations may add other fields according to their industry and quality-management requirements.

3.2 Sampling Plan

The Control Plan should clearly specify how data will be collected.

For example:

Characteristic: Customer response time

Sample size: 20 transactions

Frequency: Daily

Measurement: Time from customer request receipt to first qualified response

Control method: I-MR chart

This is much stronger than simply stating:

“Monitor response time.”

The Control Plan must make the monitoring requirement sufficiently clear that different people can execute it consistently.

3.3 Reaction Plan

A Control Plan is incomplete if it says what to measure but does not explain what to do when the result is abnormal.

A reaction plan should specify:

What constitutes an abnormal condition?
Who must be notified?
What immediate action is required?
What potentially affected output must be contained?
How should the cause be investigated?
What corrective action is required?
How is effectiveness verified?
When can normal processing resume?
What records must be completed?

3.4 Example Reaction Plan

Suppose a manufacturing process has a critical diameter requirement.

The Control Plan specifies:

Measurement: Digital micrometer
Sample: 5 pieces
Frequency: Every hour
Control method: Xbar-R chart

If a special-cause signal occurs:

Immediate Response

Stop and assess the process condition.
Notify the process owner or designated responsible person.
Identify potentially affected material.
Verify the measurement system.

Investigation

Check:

Machine settings
Tool condition
Material
Operator method
Environmental conditions
Maintenance activity

Corrective Action

Correct the verified cause.

Verification

Collect additional measurements and confirm that the process has returned to the defined operating condition.

Documentation

Record:

Date/time
Condition detected
Investigation
Cause
Corrective action
Verification result
Responsible person

This converts SPC from passive monitoring into active process management.

3.5 Preventive vs Detective Controls

A useful Control Plan identifies whether a control is:

Preventive

Designed to stop an error from occurring.

Example:

An automated system prevents a transaction from being submitted without a mandatory field.

Detective

Designed to identify an error after it occurs.

Example:

A daily audit checks completed transactions for missing information.

Preferred Approach

Where technically and economically feasible, prevention is generally preferable to relying exclusively on detection.

3.6 Control Plan and FMEA

The Control Plan should be connected to risk analysis.

A common relationship is:

FMEA → Identify Risks → Identify Critical Characteristics → Define Controls → Control Plan

FMEA helps identify:

Potential failure modes
Effects
Causes
Existing controls
Risk priorities

The Control Plan translates important controls into operational monitoring requirements.

The two documents should therefore be consistent.

If a significant process risk is identified in the FMEA but there is no corresponding control or reaction method where appropriate, the organization should review the gap.

3.7 Control Plan and Standard Work

The Control Plan describes what must be controlled.

Standard work describes how the work should be performed.

For example:

Control Plan:
Verify machine temperature every two hours.

Standard Work:

Read the temperature display.
Record the value.
Compare against the defined operating range.
If abnormal, follow the reaction procedure.

These documents should support one another rather than contradict each other.

3.8 Process Ownership

Sustaining improvement requires clear ownership.

The Green Belt may lead the improvement project, but the Process Owner normally becomes responsible for ongoing operational control.

Ownership should include:

Monitoring performance
Reviewing control-chart signals
Ensuring reaction plans are followed
Maintaining standard work
Training personnel
Escalating significant problems
Updating documentation when processes change

The project should not remain dependent on the Green Belt indefinitely.

3.9 Visual Management

Visual management can make process control easier.

Examples include:

Performance boards
Daily KPI displays
Control charts
Andon indicators
Status dashboards
Color-coded work instructions
Escalation displays

The purpose is to make abnormal conditions visible quickly.

Visual management should support decision-making rather than simply create additional displays.

3.10 Process Monitoring

Process monitoring involves systematically reviewing performance after improvement.

Monitoring may include:

Control charts
KPIs
Defect rates
Cycle times
Yield
Customer complaints
Process capability
Audit results
First-pass yield
Rework
Scrap

Monitoring frequency should reflect the risk and behavior of the process.

A highly sensitive process may require frequent monitoring.

A stable low-risk process may require less frequent review.

4. Tools / Methodology

4.1 Developing a Control Plan

A Green Belt can use the following methodology.

Step 1 — Identify the Process

Define the process to be controlled.

Examples:

Invoice processing
Machining
Customer complaint handling
Loan approval
Order fulfillment

Step 2 — Identify Critical Characteristics

Determine which outputs and inputs are important.

Consider:

CTQs
Customer requirements
Safety
Regulatory requirements
High-risk failure modes
Significant process parameters

Step 3 — Define the Standard

Specify:

Target
Specification
Operating range
Acceptance criteria
Required condition

Step 4 — Define Measurement

Specify:

What is measured?
How?
With what equipment?
By whom?
Using what operational definition?

Step 5 — Define Sampling

Specify:

Sample size
Sampling frequency
Sampling location
Sampling method

Step 6 — Select Control Method

Possible methods include:

Standard work
Checklist
Error-proofing
Inspection
SPC
Automated monitoring
Preventive maintenance
Audit

Step 7 — Define Reaction Plan

Clearly state what happens when the process deviates.

Step 8 — Assign Responsibility

Identify:

Operator
Supervisor
Process Owner
Quality representative
Maintenance
Other responsible functions

Step 9 — Establish Records

Define where evidence of control is maintained.

Step 10 — Review and Update

The Control Plan should be reviewed when:

Process changes occur.
Equipment changes.
Materials change.
Customer requirements change.
New failure modes are identified.
Significant process problems occur.
Corrective actions alter the process.

4.2 Example Control Plan

Consider an order-processing process.

Process Objective

Reduce order-entry errors.

Process Step	Characteristic	Standard	Measurement	Frequency	Control	Reaction
Receive order	Required fields	100% complete	System validation	Every order	Automated validation	Return incomplete order
Enter order	Product code	Correct code	System check	Every order	Error-proofing	Correct before release
Review order	Critical information	Accurate	Checklist	Every order	Standard work	Correct discrepancy
Release order	Order accuracy	≥ 99.5%	Audit	Daily	p chart / KPI	Investigate abnormal trend
Process performance	Error rate	Within defined target	KPI/SPC	Weekly	Dashboard	Escalation and RCA

This illustrates how individual controls combine into a system of process control.

4.3 Minitab and Process Monitoring

Where SPC is appropriate, Minitab can be used to create and monitor control charts.

Typical workflow:

Prepare data.
Confirm the measurement definition.
Select the appropriate control chart.
Enter subgroup information if applicable.
Generate the chart.
Review control limits.
Investigate special-cause signals.
Document actions.
Continue monitoring.

The software supports analysis, but it does not replace process knowledge.

5. Worked Example / Case Study

Case Study: Customer Service Response Time

A service organization completed a Six Sigma project to reduce customer response time.

Before Improvement

Average response time:

24 hours

The process showed substantial variation.

Improvement

The team:

Standardized work allocation.
Introduced a priority classification system.
Created response-time standards.
Added automated notifications.
Established escalation rules.

After Improvement

Average response time:

8 hours

The process is now stable and meets the defined performance requirement.

The project team must now determine how to sustain the improvement.

Step 1 — Identify CTQ

Customer response time is a CTQ.

Step 2 — Define the Operational Measure

Response time is defined as:

Time from receipt of a valid customer request to the first qualified response.

This prevents different employees from measuring different things.

Step 3 — Define the Standard

Target:

≤ 8 hours

Additional internal thresholds may be established according to business requirements.

Step 4 — Define Monitoring

The process owner reviews:

Daily response-time performance.
Weekly trend.
Percentage exceeding the requirement.
Special-cause signals where SPC is appropriate.

Step 5 — Establish Responsibility

Customer-service supervisors are responsible for daily monitoring.

The Process Owner reviews weekly performance.

Management receives escalation when predefined thresholds are exceeded.

Step 6 — Define Reaction Plan

If performance exceeds the defined threshold:

Identify affected transactions.
Determine whether the issue is isolated or systemic.
Check staffing and workload.
Review priority classification.
Check system availability.
Review adherence to standard work.
Identify the cause.
Take corrective action.
Verify recovery.
Record the event.

Step 7 — Sustainment

The improved process is incorporated into:

Standard work
Training
Control Plan
Performance monitoring
Management review

The Green Belt can then formally transition ownership to the process owner.

6. Practical Application

Exercise 1 — Create a Control Plan

Select a process you understand.

Examples:

Production
Purchasing
Customer service
Healthcare administration
Banking
Logistics
Education
IT service

Identify at least five important characteristics.

For each characteristic define:

Process step
Characteristic
Requirement
Measurement method
Sample size
Frequency
Control method
Responsible person
Reaction plan

Exercise 2 — Develop a Reaction Plan

Consider this situation:

A process normally operates with a defect rate below 2%.

The control chart indicates an unusual increase.

Develop a reaction plan answering:

Who should be notified?
What should happen immediately?
What output should be contained?
What data should be reviewed?
What potential causes should be investigated?
Who approves corrective action?
How will recovery be verified?
What records should be maintained?

Exercise 3 — Preventive vs Detective Controls

Classify each control as primarily preventive or detective.

A.

An automated system prevents submission when a mandatory field is blank.

Preventive

B.

A supervisor audits completed transactions every Friday.

Detective

C.

A machine interlock prevents operation when a safety guard is open.

Preventive

D.

Final inspection identifies defective products.

Detective

Discuss which controls could potentially be redesigned to prevent the error earlier in the process.

Exercise 4 — Control Plan Review

Review an existing Control Plan and ask:

Are all CTQs covered?
Are critical process inputs controlled?
Are specifications clear?
Are measurement methods defined?
Is the measurement system adequate?
Is sampling appropriate?
Are reaction plans specific?
Are responsibilities clear?
Is the FMEA aligned?
Is standard work aligned?
Are records maintained?
Is the plan current?

6.1 Control Plan Implementation Checklist

Before approving a Control Plan, verify:

Process steps are clearly defined.
Critical characteristics are identified.
CTQs are identified.
Requirements are clearly stated.
Measurement methods are defined.
Measurement systems are appropriate.
Sample size is defined.
Sampling frequency is defined.
Control methods are appropriate.
Preventive controls are considered.
SPC is used where appropriate.
Reaction plans are clearly documented.
Responsibilities are assigned.
Records are identified.
FMEA and Control Plan are aligned.
Standard work is aligned.
Process ownership is established.
Review frequency is defined.
Change-management requirements are defined.

7. Lesson Summary

A Control Plan converts a successful process improvement into a structured system for maintaining performance.

The major principles are:

Improvement must be sustained after project completion.
A Control Plan defines how important process characteristics will be controlled.
CTQs and critical process characteristics should receive appropriate attention.
Measurement methods and operational definitions must be clear.
Sampling size and frequency should be defined.
Control methods should be appropriate to the risk and process.
Prevention should be considered in addition to detection.
Reaction plans are essential.
Control Plans should align with FMEA and standard work.
Process owners must understand and accept their responsibilities.
SPC can provide ongoing statistical monitoring where appropriate.
Control Plans should be reviewed whenever significant process changes occur.
Sustaining improvement is an operational responsibility, not merely a project-team activity.

The basic Control Plan logic is:

What to Control → How to Measure → How Often → Who Controls → What to Do When Abnormal

8. Lesson Learnt / Conclusion

The Control phase of DMAIC is where improvement becomes part of normal business operations.

A Six Sigma project should not depend permanently on the Green Belt or project team. The improved process must be transferred into the organization through:

Standard work
Training
Process ownership
Monitoring
Control Plans
Reaction plans
Performance reviews
Appropriate escalation

The most important question is not:

“Did we improve the process?”

It is:

“How will we know that the improvement continues?”

A well-designed Control Plan provides the answer.

The Green Belt should therefore ensure that every significant improvement has a clear method for measurement, monitoring, response, ownership, and review.

The ultimate objective is to make the improved performance the new normal, rather than a temporary result achieved during the project.

Key takeaway:

An improvement is not fully successful until it can be sustained without depending on the project team.