fmea-facilitator

name	fmea-facilitator
description	Failure Mode and Effects Analysis facilitation skill for risk identification and prioritization.
allowed-tools	Bash(*) Read Write Edit Glob Grep WebFetch
metadata	{"author":"babysitter-sdk","version":"1.0.0","category":"quality-engineering","backlog-id":"SK-IE-018"}
graph	{"domains":["domain:industrial-engineering"],"skillAreas":["skill-area:statistical-analysis","skill-area:organizational-design","skill-area:data-analysis"],"roles":["role:operations-analyst","role:research-engineer"]}

fmea-facilitator

You are fmea-facilitator - a specialized skill for facilitating Failure Mode and Effects Analysis for risk identification and prioritization.

Overview

This skill enables AI-powered FMEA including:

• FMEA scope and boundary definition
• Failure mode brainstorming facilitation
• Severity, Occurrence, Detection rating guidance
• RPN (Risk Priority Number) calculation
• AIAG-VDA Action Priority (AP) methodology
• Control plan integration
• Recommended action tracking
• FMEA revision and living document management

Capabilities

1. FMEA Structure

from dataclasses import dataclass, field
from typing import List, Optional
from datetime import datetime
from enum import Enum

class FMEAType(Enum):
    DESIGN = "DFMEA"
    PROCESS = "PFMEA"
    SYSTEM = "SFMEA"

@dataclass
class FailureMode:
    id: str
    item_function: str
    potential_failure_mode: str
    potential_effects: List[str]
    severity: int  # 1-10
    potential_causes: List[str]
    occurrence: int  # 1-10
    current_controls_prevention: List[str]
    current_controls_detection: List[str]
    detection: int  # 1-10
    recommended_actions: List[str] = field(default_factory=list)
    responsibility: str = ""
    target_date: Optional[datetime] = None
    actions_taken: str = ""
    severity_after: Optional[int] = None
    occurrence_after: Optional[] = 
    detection_after: [] = 


     ():
         .severity * .occurrence * .detection


     ():
         ([.severity_after, .occurrence_after, .detection_after]):
             .severity_after * .occurrence_after * .detection_after
         


 :
    fmea_type: FMEAType
    item_name: 
    revision: 
    team_members: []
    start_date: datetime
    failure_modes: [FailureMode] = field(default_factory=)

     ():
        .failure_modes.append(fm)

     ():
         [fm  fm  .failure_modes  fm.rpn >= threshold]

2. Severity Rating Guide

SEVERITY_RATINGS = {
    10: {
        "effect": "Hazardous without warning",
        "description": "Very high severity ranking when potential failure mode affects safe operation without warning",
        "criteria": "May endanger operator; Noncompliance with regulations"
    },
    9: {
        "effect": "Hazardous with warning",
        "description": "Very high severity ranking when potential failure mode affects safe operation with warning",
        "criteria": "May endanger operator with warning"
    },
    8: {
        "effect": "Very High",
        "description": "Product inoperable, loss of primary function",
        "criteria": "100% of product affected, customer very dissatisfied"
    },
    7: {
        "effect": "High",
        "description": "Product operable but performance level reduced",
        "criteria": "Most customers dissatisfied"
    },
    6: {
        "effect": "Moderate",
        "description": "Product operable but comfort/convenience items inoperable",
        "criteria": "Customer experiences discomfort"
    },
    5: {
        "effect": "Low",
        "description": "Product operable but comfort/convenience items reduced",
        "criteria": "Customer somewhat dissatisfied"
    },
    : {
        : ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    }
}

 ():
    
     SEVERITY_RATINGS

3. Occurrence Rating Guide

OCCURRENCE_RATINGS = {
    10: {"probability": "Very High", "rate": ">= 100 per 1000", "cpk": "< 0.33"},
    9: {"probability": "High", "rate": "50 per 1000", "cpk": ">= 0.33"},
    8: {"probability": "High", "rate": "20 per 1000", "cpk": ">= 0.51"},
    7: {"probability": "Moderately High", "rate": "10 per 1000", "cpk": ">= 0.67"},
    6: {"probability": "Moderate", "rate": "5 per 1000", "cpk": ">= 0.83"},
    5: {"probability": "Moderate", "rate": "2 per 1000", "cpk": ">= 1.00"},
    4: {"probability": "Moderately Low", "rate": "1 per 1000", "cpk": ">= 1.17"},
    3: {"probability": "Low", "rate": "0.5 per 1000", : },
    : {: , : , : },
    : {: , : , : }
}

4. Detection Rating Guide

DETECTION_RATINGS = {
    10: {
        "likelihood": "Almost Impossible",
        "description": "No known controls to detect failure mode",
        "criteria": "Cannot detect or not checked"
    },
    9: {
        "likelihood": "Very Remote",
        "description": "Controls probably will not detect",
        "criteria": "Control achieved with indirect or random checks only"
    },
    8: {
        "likelihood": "Remote",
        "description": "Controls have poor chance of detection",
        "criteria": "Control achieved with visual inspection only"
    },
    7: {
        "likelihood": "Very Low",
        "description": "Controls have poor chance of detection",
        "criteria": "Control achieved with double visual inspection"
    },
    6: {
        "likelihood": "Low",
        "description": "Controls may detect",
        "criteria": "Control achieved with charting methods (SPC)"
    },
    5: {
        "likelihood": "Moderate",
        "description": "Controls may detect",
        "criteria": "Control based on variable gauging after parts leave station"
    },
    4: {
        "likelihood": ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    },
    : {
        : ,
        : ,
        : 
    }
}

5. AIAG-VDA Action Priority (AP)

def calculate_action_priority(severity, occurrence, detection):
    """
    Calculate Action Priority per AIAG-VDA FMEA handbook

    Returns: 'H' (High), 'M' (Medium), 'L' (Low)
    """
    # High Priority
    if severity >= 9 and occurrence >= 4:
        return "H"
    if severity >= 9 and detection >= 7:
        return "H"
    if severity >= 5 and occurrence >= 7 and detection >= 5:
        return "H"

    # Low Priority
    if severity <= 4 and occurrence <= 3:
        return "L"
    if severity <= 4 and detection <= 3:
        return "L"
    if severity <= 6 and occurrence <= 2 and detection <= 4:
        return "L"

    # Medium Priority (default)
    return "M"

def get_ap_table():
    """
    Return AP lookup table summary
    """
     {
        : {
            : ,
            : ,
            : 
        },
        : {
            : ,
            : ,
            : 
        },
        : {
            : ,
            : ,
            : 
        }
    }

6. FMEA Analysis and Reporting

def analyze_fmea(fmea: FMEA):
    """
    Analyze FMEA and generate summary report
    """
    analysis = {
        "total_failure_modes": len(fmea.failure_modes),
        "rpn_statistics": {},
        "ap_distribution": {"H": 0, "M": 0, "L": 0},
        "top_risks": [],
        "actions_status": {
            "pending": 0,
            "in_progress": 0,
            "completed": 0
        }
    }

    rpns = [fm.rpn for fm in fmea.failure_modes]
    if rpns:
        analysis["rpn_statistics"] = {
            "max": max(rpns),
            "min": min(rpns),
            "average": sum(rpns) / len(rpns),
            "above_100": sum(1 for r in rpns if r >= 100),
            "above_200": sum(1 for r in rpns if r >= 200)
        }

    # AP distribution
    for fm in fmea.failure_modes:
        ap = calculate_action_priority(fm.severity, fm.occurrence, fm.detection)
        analysis[][ap] += 

    
    sorted_fms = (fmea.failure_modes, key= x: x.rpn, reverse=)
    analysis[] = [
        {
            : fm.,
            : fm.potential_failure_mode,
            : fm.rpn,
            : fm.severity,
            : fm.occurrence,
            : fm.detection,
            : calculate_action_priority(fm.severity, fm.occurrence, fm.detection)
        }
         fm  sorted_fms[:]
    ]

    
     fm  fmea.failure_modes:
         fm.actions_taken:
            analysis[][] += 
         fm.recommended_actions:
            analysis[][] += 
        :
            analysis[][] += 

     analysis

 ():
    
    control_items = []

     fm  fmea.failure_modes:
         control  fm.current_controls_prevention + fm.current_controls_detection:
            control_items.append({
                : fm.item_function,
                : fm.potential_failure_mode,
                : control,
                : ,
                : ,
                : 
            })

     control_items

Process Integration

This skill integrates with the following processes:

• failure-mode-effects-analysis.js
• root-cause-analysis-investigation.js
• statistical-process-control-implementation.js

Output Format

{
  "fmea_summary": {
    "type": "PFMEA",
    "item": "Assembly Process",
    "revision": "1.2",
    "total_failure_modes": 45
  },
  "risk_analysis": {
    "high_priority_count": 8,
    "medium_priority_count": 22,
    "low_priority_count": 15,
    "max_rpn": 392,
    "avg_rpn": 112
  },
  "top_risks": [
    {
      "failure_mode": "Missing fastener",
      "rpn": 392,
      "ap" 
       
    
  
   

Best Practices

1. Cross-functional team - Include design, manufacturing, quality, service
2. Start early - Begin FMEA during design phase
3. Focus on prevention - Prioritize prevention over detection
4. Living document - Update FMEA when process changes
5. Use AP not just RPN - Consider AIAG-VDA AP methodology
6. Track actions - Close the loop on recommended actions

Constraints

• Document all assumptions
• Use consistent rating criteria
• Review periodically
• Link to control plans