FMEA in Lean Six Sigma Yellow Belt Projects Key Elements RPN and Practical Applications
A defect is rarely a surprise after the fact. Once a part is scrapped, a shipment is late, or a customer complaint arrives, the warning signs often become clear. Failure Mode and Effects Analysis, better known as FMEA, gives manufacturing teams a structured way to spot those warning signs before they turn into problems.
In a Lean Six Sigma Yellow Belt project, FMEA is especially useful because it connects process knowledge with risk-based thinking. Yellow Belt projects often focus on practical improvements, such as reducing scrap, improving first-pass yield, shortening changeover time, or preventing rework. FMEA supports those goals by asking a simple but powerful question:
What could go wrong, how bad would it be, how likely is it, and how well can we catch it before it reaches the next step or the customer?
The answer helps teams focus their limited time and effort on the risks that matter most.
What FMEA means in a Yellow Belt project
FMEA is a structured method for identifying possible failures in a product, process, or service. In manufacturing, a process FMEA usually looks at each step in production and asks what could fail at that step.
A basic process FMEA may include:
The process step being reviewed
The potential failure mode
The potential effect of the failure
The possible cause
Current controls
Severity rating
Occurrence rating
Detection rating
Risk Priority Number
Recommended actions
Owner and due date
Updated ratings after action
For a Yellow Belt project, the goal is not to create a complex document for its own sake. The goal is to make risk visible and guide better decisions. A simple FMEA, built with input from operators, technicians, supervisors, quality staff, and process owners, can prevent teams from chasing low-impact issues while serious risks remain untreated.
FMEA fits well inside the DMAIC cycle, which stands for Define, Measure, Analyze, Improve, and Control. It is often most useful in the Analyze and Improve phases, when the team is trying to understand causes and choose the right countermeasures. It can also support the Control phase by confirming that new controls reduce process risk.
The three key elements of FMEA
FMEA uses three core ratings to describe risk: Severity, Occurrence, and Detection. These ratings are usually scored on a scale from 1 to 10. The exact definitions should be agreed on by the team or based on company standards, but the concept stays the same.
FMEA element | What it measures | Low score means | High score means |
Severity | Impact if the failure happens | Little or no effect | Serious safety, quality, customer, or compliance impact |
Occurrence | Likelihood that the cause will happen | Rare or unlikely | Frequent or expected |
Detection | Ability to catch the failure before it escapes | Very likely to detect | Hard to detect or no control exists |
These three values form the basis of the Risk Priority Number.
Severity measures the impact of failure
Severity asks, “If this failure occurs, how serious is the effect?”
In a manufacturing process, severity can relate to:
Operator safety
Customer safety
Product function
Regulatory compliance
Scrap and rework
Delivery performance
Customer satisfaction
A low severity rating might apply to a minor cosmetic issue that does not affect fit, function, safety, or customer use. A high severity rating might apply to a brake component made to the wrong dimension, a missing weld on a load-bearing part, or incorrect labeling on a regulated product.
Severity is different from probability. A failure may be extremely rare but still severe if the result is dangerous or costly. For this reason, teams should avoid lowering severity just because the issue does not happen often. Severity reflects the effect, not the chance.
Occurrence measures the likelihood of the cause
Occurrence asks, “How likely is the cause of this failure to happen?”
In practice, occurrence is tied to the process cause, not just the defect. For example, a missing fastener may be the failure mode. The cause might be skipped manual assembly, feeder jam, incorrect work instruction, or lack of torque confirmation.
Occurrence can be estimated using:
Historical defect data
Scrap and rework records
Maintenance logs
Audit findings
Operator feedback
Process capability data
Supplier performance records
A low occurrence rating means the cause is rare or well controlled. A high occurrence rating means the cause happens often, or the team has evidence that it could happen under normal operating conditions.
For a Yellow Belt project, occurrence ratings do not need to be perfect. They need to be reasonable, consistent, and based on the best available process knowledge.
Detection measures the strength of current controls
Detection asks, “If the failure happens, how likely are current controls to catch it before it moves forward?”
This rating often causes confusion because the scale feels backward. A low detection score is good because the team is confident the failure will be caught. A high detection score is risky because current controls are weak, manual, inconsistent, or missing.
Strong detection controls may include:
Automated sensors with stop functions
Error-proofing devices that prevent incorrect assembly
Vision systems that reject nonconforming parts
In-process tests that verify critical dimensions
Barcode scans that prevent wrong-part use
Weak detection controls may include:
Visual checks performed under time pressure
End-of-line inspection only
Random audits with small sample sizes
Reliance on memory or experience
No inspection before shipment
Detection should describe how well the process catches the problem today, before new improvements are added.
How Severity Occurrence and Detection create the RPN
The Risk Priority Number, or RPN, is calculated by multiplying the three ratings:
RPN = Severity × Occurrence × Detection
If each rating uses a 1 to 10 scale, the RPN can range from 1 to 1,000.
For example:
Failure mode | Severity | Occurrence | Detection | RPN |
Incorrect label applied to finished part | 6 | 4 | 5 | 120 |
Missing torque on critical fastener | 9 | 3 | 6 | 162 |
Minor surface scratch on hidden area | 2 | 5 | 3 | 30 |
Wrong component loaded into feeder | 8 | 4 | 8 | 256 |
The wrong component issue has the highest RPN in this example because it has a serious effect, a moderate chance of happening, and weak detection. That makes it a strong candidate for action.
RPN helps teams compare different risks, but it should not be used blindly. A failure mode with very high severity may require action even if the RPN is not the highest number on the list. For example, a safety-related failure rated high on severity deserves attention even when occurrence is low.
A practical Yellow Belt approach is to use RPN as a guide, then apply judgment. Look for:
High RPN values
High severity ratings
Failure modes with poor detection
Quick fixes that reduce risk with little cost
Repeated issues that create waste or customer complaints
After the team completes corrective actions, the FMEA should be updated. The new ratings show whether the change reduced risk. In many cases, actions reduce occurrence or detection. Severity often stays the same because the effect of the failure has not changed.
Examples of FMEA in manufacturing processes
FMEA becomes clearer when applied to real production scenarios. The examples below show how it can identify risks that may otherwise stay hidden until a defect appears.
Example one shows a packaging process with wrong labels
A manufacturer packs similar-looking products in the same area. Operators select labels from a nearby rack and apply them before boxing. Customer complaints show that wrong labels occasionally reach the field.
The FMEA might identify:
Process step | Potential failure mode | Potential effect | Possible cause | Current control |
Apply product label | Wrong label applied | Customer receives incorrect product information | Similar labels stored together | Operator visual check |
Pack finished goods | Wrong box used | Shipment mismatch | Boxes look similar | End-of-line inspection |
Print batch label | Incorrect date code | Traceability issue | Manual entry error | Supervisor review |
The RPN may show that wrong label application is the top risk because detection relies on visual checking. Recommended actions could include barcode verification, label color separation, locked label bins, or software that prints labels only after scanning the work order.
Example two shows an assembly process with missing fasteners
In a manual assembly cell, an operator installs four fasteners. A missing fastener can cause noise, vibration, or product failure during use.
The FMEA might reveal several possible causes:
Fastener bin runs empty during the cycle
Operator gets interrupted
Work instruction is unclear
No torque confirmation exists
Final inspection cannot see the hidden fastener location
The team may give the failure a high severity score because the product can fail in use. Occurrence may be moderate if missing fasteners have appeared in defect logs. Detection may be poor if the fastener is hidden after assembly.
Good actions might include a torque tool with count verification, a fixture that will not release the part until all fasteners are torqued, or a parts-present sensor. These changes reduce the chance that the failure occurs or escapes.
Example three shows a machining process with incorrect dimensions
A machining operation produces a shaft with a critical diameter. Tool wear can slowly cause the diameter to drift out of specification.
The FMEA could identify:
Failure mode Diameter out of tolerance
Effect Bearing fit fails during assembly or field use
Cause Tool wear, incorrect offset, or skipped first-piece inspection
Current control Periodic operator measurement
If the measurement frequency is too low, detection may be weak. The action plan may include tool life limits, automatic offset controls, first-piece approval after each tool change, and control charts for the critical dimension.
This is where FMEA connects closely with Lean Six Sigma. The team can use data to understand variation, then use mistake-proofing and control plans to reduce risk.
Practical uses of FMEA across industries
Although FMEA is common in manufacturing, its value extends far beyond one type of factory. Any process with repeatable steps, quality requirements, and potential failure points can benefit from it.
Automotive and transportation
Automotive suppliers use FMEA to manage risks in machining, welding, painting, assembly, electronics, and safety-critical components. It supports quality planning by identifying how parts could fail and what controls should prevent escapes.
For example, a supplier may use FMEA to study weld strength, connector seating, torque application, or part traceability. High-risk items often receive stronger controls such as automated inspection, controlled fixtures, or process alarms.
Aerospace and defense
Aerospace processes require strict control because failures can have serious consequences. FMEA helps teams examine risks in material handling, special processes, assembly, documentation, and inspection.
In this setting, severity ratings often drive action. Even low-frequency risks may receive attention if the effect could compromise safety, mission performance, or compliance.
Medical device and pharmaceutical manufacturing
Medical device and pharmaceutical teams use FMEA to review process steps where contamination, labeling errors, incorrect dosage, equipment failure, or packaging defects could affect users. The method supports risk management and process control.
Examples include checking sterile barrier sealing, part mix-ups, lot traceability, and test equipment setup. Detection controls may include validated inspection systems, line clearance checks, and electronic records.
Food and beverage production
Food and beverage operations can use FMEA to identify risks related to allergen control, fill levels, seal integrity, foreign material, sanitation, and labeling. A wrong label can be especially serious when allergens are involved.
FMEA helps teams see where simple process changes, such as separate storage, scan verification, or clearer line clearance rules, can reduce risk.
Electronics manufacturing
Electronics processes include many steps where defects can be hard to see. FMEA can review solder bridging, polarity errors, missing components, electrostatic discharge risk, incorrect firmware loading, and test coverage.
Because some defects are not visible, detection ratings can be high unless automated test systems or functional checks are in place.
Service and administrative processes
FMEA also works in nonproduction processes. A purchasing team can use it to study supplier data errors. A maintenance team can use it to review preventive maintenance failures. A warehouse team can use it to reduce picking errors.
The same logic applies. Define the process step, identify what could go wrong, rate the risk, and act on the most serious issues.
Benefits of using FMEA in Lean Six Sigma Yellow Belt work
FMEA gives Yellow Belt projects a practical way to move from opinion to structured analysis. It does not replace data, process maps, root cause analysis, or control plans. It helps connect them.
Key benefits include:
Better prioritization Teams can focus on the risks with the strongest mix of impact, likelihood, and weak detection.
Earlier problem prevention FMEA encourages prevention before defects reach inspection, rework, or customers.
Stronger cross-functional input Operators, maintenance staff, quality teams, and supervisors often see different risks. FMEA brings that knowledge together.
Clearer control plans The output of an FMEA can guide inspection points, mistake-proofing, standard work, and monitoring.
Reduced waste Preventing failure reduces scrap, rework, sorting, downtime, and emergency response.
Improved customer confidence A process that identifies and manages risk is more likely to deliver consistent quality.
For Yellow Belt practitioners, FMEA also builds problem-solving discipline. It teaches teams to ask better questions and support decisions with evidence.
How to keep an FMEA useful
An FMEA should be a working document, not a file that gets completed once and forgotten. To keep it useful, make it simple enough for the team to maintain.
A good practice is to review the FMEA when:
A customer complaint occurs
A new defect trend appears
Equipment or tooling changes
A supplier change occurs
A process step changes
A new product is launched
Corrective actions are completed
The best FMEAs are specific. “Operator error” is not a strong cause because it does not point to a fix. Better causes include unclear work instruction, missing fixture stop, similar-looking parts, no scan verification, worn locating pin, or skipped setup check.
The action plan should also be specific. “Train operator” may help, but it is often a weak long-term control if the process still allows the mistake. Stronger controls change the process so the error is less likely or easier to catch.
For example, if two parts can be mixed up, better lighting and retraining may help for a while. A stronger action may be a keyed fixture that accepts only the correct part, or barcode scanning that stops the process when the wrong part is presented.
A practical takeaway for Yellow Belt teams
FMEA gives Lean Six Sigma Yellow Belt projects a clear way to identify and reduce process risk. Severity shows the impact of a failure. Occurrence shows how likely the cause is. Detection shows how well current controls can catch it. When multiplied together, they create the RPN, which helps teams decide where to act first.
The real value is not the number by itself. The value comes from the conversation, the process knowledge, and the improvements that follow. A well-built FMEA helps teams prevent defects, protect customers, reduce waste, and strengthen day-to-day control.
Start with one process step, one known problem area, or one customer complaint. Map the possible failures, rate them honestly, and act on the risks that matter most. That is where FMEA becomes more than a form. It becomes a practical tool for better manufacturing performance.
Comments