FMEA Analysis

Failure Mode and Effects Analysis (FMEA) is a systematic method for identifying potential failures in a product, process, or system and prioritizing them by risk. ProBeya provides a full FMEA workflow: from creating analyses and scoring failure modes through recording corrective actions and measuring risk reduction with Pareto charts.

FMEA is critical in pharma and manufacturing environments where unmitigated process failures can lead to product defects, safety incidents, or regulatory non-compliance.

Overview

When a process or product has known quality risks, teams need a structured way to catalog those risks, assess their severity, and focus improvement efforts where they matter most. Without prioritization, teams either fix easy problems (low impact) or try to fix everything at once (resource exhaustion).

FMEA solves this with the Risk Priority Number (RPN) -- a composite score of Severity, Occurrence, and Detection that objectively ranks failure modes. ProBeya computes the RPN server-side to prevent inconsistencies, and provides a Pareto chart to visualize where 20% of failure modes account for 80% of total risk.

Getting Started

1. Navigate to a workspace and select Lean > FMEA from the sidebar.
2. Click New FMEA Analysis and provide a title, process name, FMEA type, and owner.
3. Assign team members who will participate in the analysis.
4. Save the analysis (starts in draft status).
5. Add failure mode items, scoring each for Severity, Occurrence, and Detection (1-10 scale).
6. Review the auto-computed RPN values and the Pareto chart to prioritize actions.
7. After implementing corrective actions, use Record Action to re-score and measure risk reduction.

Key Concepts

Term	Definition
Failure Mode	A specific way in which a process step, product component, or system can fail.
Severity (S)	How serious the effect of the failure would be (1 = negligible, 10 = catastrophic).
Occurrence (O)	How likely the failure is to happen (1 = extremely unlikely, 10 = almost certain).
Detection (D)	How likely current controls will detect the failure before it reaches the customer (1 = certain detection, 10 = no detection).
RPN	Risk Priority Number = S x O x D. Range: 1-1000. Higher values indicate higher risk.
Corrective Action	A change implemented to reduce the RPN by improving severity, occurrence, or detection scores.
Post-Action RPN	The new RPN after a corrective action has been implemented and re-scored. The delta between original and post-action RPN quantifies risk reduction.
Pareto Chart	A bar chart showing failure modes ordered by RPN with a cumulative percentage line, revealing which few items account for the majority of risk.

How It Works

FMEA Types

ProBeya supports multiple FMEA types to match the analysis context:

• Process FMEA (PFMEA) -- Analyzes manufacturing or operational process steps for failure modes.
• Design FMEA (DFMEA) -- Analyzes product design elements for potential failure modes before production.
• System FMEA -- Analyzes system-level interactions and integration failure points.

Scoring and RPN Computation

When adding a failure mode item, you provide three scores on a 1-10 scale:

1. Severity -- Impact if the failure occurs
2. Occurrence -- Likelihood of the failure occurring
3. Detection -- Ability of current controls to catch the failure

The RPN is always computed server-side as Severity x Occurrence x Detection. The client never sends a pre-computed RPN value. This prevents score manipulation and ensures mathematical consistency.

Items are displayed sorted by RPN descending (highest risk first), which is the standard FMEA presentation order for prioritization.

Recording Corrective Actions

After the team implements a corrective action for a failure mode, the recordAction procedure captures:

• Action taken -- Description of what was done
• New Severity, Occurrence, Detection -- Re-scored values after the action
• New RPN -- Computed server-side from the new scores

The item status transitions to verified, indicating the action has been taken and its effectiveness evaluated. The difference between the original RPN and the new RPN is the quantitative proof of FMEA effectiveness.

Pareto Analytics

The getRpnPareto endpoint returns failure modes ordered by RPN descending with two additional computed fields:

• Percentage -- Each item's share of total risk (item RPN / total RPN x 100)
• Cumulative Percentage -- Running total of percentages

This data structure powers the Pareto chart visualization, making it immediately visible which handful of failure modes contribute the majority of risk -- the 80/20 principle applied to risk management.

Configuration

FMEA analyses are workspace-scoped. Each analysis can optionally link to a board (project) and a problem sheet.

Setting	Description
FMEA Type	Process, Design, or System -- determines the analysis context.
Team Members	Array of user IDs participating in the analysis. Stored as JSONB for flexible team composition.
Status	Draft, in-progress, completed -- tracks the analysis lifecycle.
Linked Problem Sheet	Optional link to a Kaizen problem sheet for DMAIC traceability.

Permissions

Role	Capabilities
Workspace Member	View analyses, add and score failure mode items, record actions
Workspace Admin	All member capabilities plus create, delete, and manage analysis status
Organization Admin	Full access across all workspaces

All mutations enforce multi-tenant isolation via organizationId from the authenticated session context.

Tips & Best Practices

Focus on the top Pareto items first

Use the Pareto chart to identify the top 3-5 failure modes contributing the most risk. Directing corrective actions at these items delivers the greatest risk reduction per unit of effort.

Score with the team, not alone

FMEA scoring is inherently subjective. Always score Severity, Occurrence, and Detection as a cross-functional team to reduce individual bias and capture diverse perspectives.

Re-score after every action

Always use the Record Action workflow after implementing a corrective action. The pre/post RPN comparison is the objective evidence of improvement that auditors and management need to see.

Use Detection score wisely

Teams often overlook the Detection score. A failure with moderate severity and occurrence but poor detection (D=9 or 10) can be more dangerous than a well-detected failure with higher severity. Improving detection is often the quickest win.

Troubleshooting

Issue	Resolution
RPN not updating after score change	RPN is recomputed automatically when any component score changes. If the displayed value seems stale, refresh the page to fetch the latest server-side computation.
Cannot add items to analysis	Verify the parent analysis belongs to your organization. The system checks ownership before allowing item creation.
Pareto chart shows all items at 0%	This occurs when total RPN is zero (all items scored as 1x1x1). Verify that realistic scores have been applied.
Record Action not available	Ensure the failure mode item exists and belongs to your organization. The Record Action mutation validates ownership before applying changes.

Related Features

• SIPOC Diagrams -- Define process boundaries before performing FMEA on individual process steps.
• Voice of Customer -- Customer needs identified in VOC studies inform which failure modes matter most (highest severity).
• Kaizen / Problem Solving -- When an FMEA reveals a critical failure mode, initiate a Kaizen project to address the root cause.
• Standard Operating Procedures -- After corrective actions are verified, update SOPs to embed the improved process controls.
• Action Log -- Track individual corrective actions assigned from FMEA reviews.