rohoon-6sigma（鲁胡恩 6sigma）

Rohoon Six Sigma Professional Support

本技能提供基于 AIAG-VDA SPC Manual (Yellow Volume) 第 1 版，2026 年 2 月、AIAG MSA Manual 第 4 版、ISO 13053 六西格玛标准 和 精益方法论 的全面质量管理支持。

🎯 Quick Navigation

Quick Start

📊 AIAG-VDA Standard Reports (NEW! ✨)

⚠️ 重要 - 报告格式稳定性

Figure 12-1/12-2/12-3 PDF 报告遵循特定的标准格式，除非用户明确指示，否则不要修改布局、字体、表格宽度或格式。

• 当前标准格式：

- 英文版：tmp/YOUR_ORIGINAL_EN.pdf (210,401 bytes) - 中文版：tmp/YOUR_ORIGINAL_v2.pdf (258,583 bytes)

• 格式基于 AIAG-VDA SPC Manual 标准
• 如需更改格式，请先询问用户
• 标准格式参考 PDF：

- 英文：references/AIAG_VDA_Standard_EN_Format.pdf - 中文：references/AIAG_VDA_Standard_ZH_Format.pdf

统一报告生成器 - 严格遵循黄皮书标准格式

# Figure 12-1: 正态分布报告 (中文)
python3 scripts/aiagvda_unified_report.py -o report.pdf -f 12-1 -l zh

# Figure 12-1: Normal Distribution Report (English)
python3 scripts/aiagvda_unified_report.py -o report.pdf -f 12-1 -l en

# Figure 12-2: 非正态/混合分布报告
python3 scripts/aiagvda_unified_report.py -o report.pdf -f 12-2 -l zh

# Figure 12-3: 标准统计表格报告
python3 scripts/aiagvda_unified_report.py -o report.pdf -f 12-3 -l zh

报告特点:

• 20-22 个编号元素 (严格匹配黄皮书格式)
• 中英文双语支持
• 几何法能力指数 (Cp,G / Cpk,G)
• 95% 置信区间
• PPM 计算
• 专业图表 (直方图、原始值图、概率图、控制图)

📊 Process Capability Analysis

# Two-sided specification
python3 scripts/calculate_capability.py \
  --data "10.1,10.2,10.15,10.18,10.12,10.05,10.22,10.08,10.14,10.19" \
  --usl 10.5 --lsl 9.5 --json

# Only upper specification
python3 scripts/calculate_capability.py \
  --data "..." --usl 10.5 --json

📈 Control Charts

# Xbar-R chart (n=5)
python3 scripts/control_chart.py \
  --data "[10.1,10.2,10.15,10.18,10.12, 10.05,10.22,10.08,10.14,10.19]" \
  --chart-type Xbar-R --subgroup-size 5 --json

# I-MR chart (individual values)
python3 scripts/control_chart.py \
  --data "10.1,10.2,10.15,10.18,10.12" \
  --chart-type I-MR --json

📏 MSA GR&R Study

# 10 parts × 3 operators × 3 trials = 90 measurements
python3 scripts/msa_grr_analysis.py \
  --data "[10.1,10.15,10.12, ...]" \
  --parts 10 --operators 3 --trials 3 \
  --tolerance 0.5 --json

🧪 DOE Full Factorial

# 2^2 full factorial design
python3 scripts/doe_full_factorial.py \
  --factors 2 \
  --responses "[40,50,45,55]" \
  --analyze

# 2^3 with center points
python3 scripts/doe_full_factorial.py \
  --factors 3 \
  --responses "[45,52,48,58,46,53,49,59]" \
  --center-points 3 \
  --analyze

# Effect calculation and visualization
python3 scripts/doe_factor_effects.py \
  --design "2^3" \
  --responses "[...]" \
  --plot main-effects interaction

🧪 Taguchi S/N Ratios

# S/N ratio calculation
python3 scripts/doe_sn_ratio.py \
  --responses "[45,52,48,...]" \
  --type larger-is-better \
  --json

📊 Response Surface

# Response Surface (CCD / Box-Behnken)
python3 scripts/doe_response_surface.py --help

📊 Excel Report Generation

# Excel report
python3 scripts/excel_report.py --help

🇨🇳 GB/T 36077-2025 中国国家标准评价框架

七大评价维度 (总分 1000 分)

DMAIC 各阶段输入、主要工作和输出

DMADV (DLSS) 各阶段输入、主要工作和输出

精益六西格玛项目评分表 (满分 110 分)

常用工具与方法 (GB/T 36077-2025 附录 B)

DMAIC 各阶段工具

• D 界定: 头脑风暴、亲和图、树图、流程图、SIPOC、平衡计分卡、水平对比、QFD、甘特图
• M 测量: 因果图、散点图、MSA、过程能力分析、因果矩阵、FTA、箱线图、直方图、抽样
• A 分析: 假设检验、方差分析、统计分布识别、时间序列分析、价值流图、FMEA、回归分析、大数据分析、机器学习
• I 改进: 试验设计、数学优化、EVOP、目视管理、5S、ECRS 原则、看板、快速换型、TOC、TRIZ
• C 控制: SPC、防错、定置管理、SOP、控制计划、OCAP

DMADV 各阶段工具

• D 界定: 顾客需求分析、市场调查、KANO 分析、QFD、多属性决策、设计计分卡
• M 测量: 关系矩阵、因果图、MSA、FMEA、过程能力分析、IPO 分析、工艺计分卡
• A 分析: 头脑风暴、FMEA、方差分析、试验设计、TRIZ、大数据分析、机器学习、FTA
• D 设计: TRIZ、DFx、公理化设计、试验设计、稳健性设计、田口方法、DFMEA、数字化仿真
• V 验证: 模拟技术、过程能力分析、假设检验、试验验证、可靠性分析、PFMEA、SPC

核心术语定义 (GB/T 36077-2025)

Core Features

1. SPC Statistical Process Control ✅

• 7 Control Charts: Xbar-R, Xbar-S, I-MR, CUSUM, EWMA, p/np, c/u
• 8 AIAG-VDA Out-of-Control Rules: Automatic detection
• 🆕 Western Electric 7 Rules: Comprehensive rule set for anomaly detection
• Process Capability: Cp, Cpk, Pp, Ppk, Pm, Pmk (Geometric Method)
• 🆕 Sigma Estimation: Range (R/d₂), Std Dev (s/c₄), Pooled Std Dev (s_p)
• Normality Tests: Shapiro-Wilk, Anderson-Darling
• 🆕 Distribution Selection: Normal, Binomial, Poisson, Weibull guidance
• 🆕 Nonparametric Control Charts: Distribution-free MEC/MCE charts (Ansari-Bradley)
• 🆕 Change-Point Detection: Single and multiple change-point monitoring
• 🆕 Multivariate SPC: MCUSUM, MEWMA, LASSO-based charts
• 🆕 Profile Monitoring: Linear and nonlinear profile analysis
• 🆕 Short Production Runs: Standardized Z-MR charts for high-mix low-volume
• 🆕 Montgomery Coefficients: A₂, D₃, D₄, d₂, c₄ tables (n=2 to 25)

2. MSA Measurement System Analysis ✅

• GR&R Studies: ANOVA method, Xbar-R method
• Bias Study: Accuracy assessment
• Linearity Study: Range consistency
• Stability Study: Time drift monitoring

3. DOE Design of Experiments ✅

• Full Factorial: 2^k designs (2-8 factors)
• Fractional Factorial: Resolution III, IV, V
• Response Surface: CCD, Box-Behnken
• Taguchi Methods: L4, L8, L9, L16, L18, L27 orthogonal arrays
• ANOVA: One-way, Two-way with interaction
• 🆕 Main & Interaction Effects: Effect calculation and visualization
• 🆕 S/N Ratios: Taguchi signal-to-noise ratio (smaller/larger/nominal-is-better)
• 🆕 Mixture Designs: Scheffé polynomial models
• 🆕 Robust Parameter Design: Control vs noise factors

4. Reports & Export ✅

• Excel Reports: Professional format with 3 worksheets
• PDF Reports: AIAG-VDA standard format
• JSON Output: For system integration

5. Advanced Analysis ✅

• EWMA Control Charts: Detect small shifts
• CUSUM Control Charts: Cumulative sum monitoring
• Multivariate Control: Hotelling T² charts
• Outlier Detection: Modified Z-score, IQR, ARIMA
• 🆕 Nonparametric MEC/MCE: Mixed EWMA-CUSUM charts (distribution-free)
• 🆕 Change-Point Detection: Mean and variance monitoring
• 🆕 Performance Metrics: RMI, AEQL, PCI, ARARL
• 🆕 Short Run SPC: Standardized charts for high-mix low-volume production
• 🆕 Capability Confidence Intervals: 95% CI for Cp/Cpk/Pp/Ppk
• 🆕 Western Electric Rules: 7 rules for out-of-control detection
• 🆕 Sigma Estimation Methods: Range, Std Dev, Pooled Std Dev
• 🆕 Distribution Selection Guide: Normal, Binomial, Poisson, Weibull

6. System Integration ✅

• HTTP API: Real-time monitoring
• ERP/MES Integration: CSV, Excel, JSON import/export
• Database Support: SQLite, PostgreSQL
• 🆕 ISA-95 B2MML: Enterprise-control system integration

Usage Examples

Define Phase

- Project Charter Template
- SIPOC Diagram
- CTQ Tree
- VOC/VOB Collection

Measure Phase

# MSA GR&R Study
python3 scripts/msa_grr_analysis.py \
  --data "[...]" --parts 10 --operators 3 --trials 3 \
  --tolerance 0.5 --json

# Baseline Capability
python3 scripts/calculate_capability.py \
  --data "..." --usl 10.5 --lsl 9.5 --json

Analyze Phase

# Control Chart with out-of-control detection
python3 scripts/control_chart.py \
  --data "[...]" --chart-type Xbar-R --subgroup-size 5

# Advanced Charts (CUSUM, EWMA, I-MR, MAMR, Hotelling T2)
python3 scripts/advanced_control_charts.py

Improve Phase

# DOE Full Factorial
python3 scripts/doe_full_factorial.py \
  --factors 3 --responses "[45,52,48,58,46,53,49,59]" \
  --analyze

# Factor Effects
python3 scripts/doe_factor_effects.py \
  --responses "[45,52,48,58,46,53,49,59]" --json

# Response Surface
python3 scripts/doe_response_surface.py --help

Control Phase

# Excel Report
python3 scripts/excel_report.py --help

# Generate test data + batch reports
python3 scripts/generate_test_data.py
python3 scripts/batch_generate_reports_v2.py

Evaluation Criteria

Process Capability

MSA GR&R

ndc (Number of Distinct Categories)

• ndc ≥ 5: ✅ Adequate resolution
• ndc < 5: ⚠️ Insufficient resolution

Standards & References

SPC Standards

• AIAG-VDA SPC Manual (Yellow Volume), 1st edition, February 2026
• ISO 22514: Statistical methods in process management
• ISO 3534: Statistics - Vocabulary and symbols
• ISO 7870: Control charts
• GB/T 36077-2025: 精益六西格玛管理评价准则 (Criteria for lean six sigma management assessment)

MSA Standards

• AIAG MSA Manual, 4th edition
• ISO 22514-7: Measurement system analysis
• VDA 5: Measurement process capability

Six Sigma Standards

• ASQ Six Sigma Body of Knowledge
• ISO 13053: Six Sigma methodology (DMAIC/DMADV)
• GB/T 36077-2025: 精益六西格玛管理评价准则 (代替 GB/T 36077-2018)

Quality System Standards

• IATF 16949: Automotive quality management
• ISO 9001: Quality management systems
• GB/T 19001: 质量管理体系 要求
• GB/T 19580: 卓越绩效评价准则

Script Tools

SPC Core

• calculate_capability.py - Process capability indices (Cp, Cpk, Pp, Ppk, Pm, Pmk)
• control_chart.py - Control charts (Xbar-R, Xbar-S, I-MR)
• attribute_control_charts.py - Attribute control charts (p/np/c/u/Z)
• advanced_control_charts.py - CUSUM, EWMA, MAMR, Hotelling T²
• pooled_std.py - Sigma estimation (Range, Std Dev, Pooled)
• western_electric_rules.py - Western Electric out-of-control rules
• spc_calculator.py - SPC calculations
• spc_report.py - SPC report generation
• excel_report.py - Excel report generation

MSA

• msa_calculator.py - MSA calculations
• msa_grr_analysis.py - GR&R study (Xbar-R method)
• msa_other_studies.py - Bias, Linearity, Stability
• msa_report.py - MSA report generation

DOE

• doe_full_factorial.py - 2^k full factorial designs
• doe_response_surface.py - CCD, Box-Behnken
• doe_factor_effects.py - Main and interaction effects
• doe_sn_ratio.py - Taguchi S/N ratio calculation
• generate_doe_report.py - DOE report generation

Utilities

• generate_test_data.py - Generate test datasets
• batch_generate_reports_v2.py - Batch PDF report generation
• data_import.py - Data import utilities
• demo_report_generation.py - Demo report generation

Version

Current: v1.8.0 (2026-04-13)

Changes in v1.8.0:

• ✅ NEW: GB/T 36077-2025 中国国家标准整合 - 完整融入《精益六西格玛管理评价准则》
• ✅ 七大评价维度框架 (领导力/顾客驱动/推进规划/项目管理/评价与激励/基础架构/实施成果)
• ✅ DMAIC/DMADV 各阶段输入 - 主要工作 - 输出完整对照表
• ✅ 精益六西格玛项目评分表 (110 分制)
• ✅ 常用工具与方法 (按 DMAIC/DMADV 各阶段分类)
• ✅ 核心术语定义 (LSS/Lean/Six Sigma/DLSS/DMAIC/DMADV/倡导者/黑带/绿带/黄带)
• ✅ 评价准则条款赋值表 (总分 1000 分)
• ✅ 基于 GB/T 36077-2025 (代替 GB/T 36077-2018),2025-12-31 发布，2026-07-01 实施
• ✅ 支持中国组织精益六西格玛管理自我评价、相关方评价、第三方评价

Changes in v1.7.0:

• ✅ NEW: Real-time data stream processing for Quality 4.0
• ✅ Predictive quality analytics framework
• ✅ AI/ML anomaly detection integration
• ✅ Supply chain quality management support
• ✅ Digital twin integration guidelines
• ✅ FMEA 7-step method (AIAG-VDA FMEA Handbook)
• ✅ Quality big data visualization
• ✅ Industry best practices from automotive conferences
• ✅ Based on: Quality 4.0 conferences, automotive industry best practices

Changes in v1.6.0:

• ✅ NEW: Western Electric 7 rules for out-of-control detection
• ✅ Sigma estimation methods (Range R/d₂, Std Dev s/c₄, Pooled s_p)
• ✅ Extended coefficient tables (d₂, c₄ for n=2 to 25)
• ✅ Distribution selection guide (Normal, Binomial, Poisson, Weibull)
• ✅ Nonparametric multivariate control charts
• ✅ Goodness-of-fit based nonparametric EWMA
• ✅ MIL-HDBK-1916 guidance
• ✅ Based on: ASQ SPC guidelines, Western Electric rules

Changes in v1.1.1:

• ✅ Fixed DOE parameter parsing
• ✅ Added JSON output for EWMA
• ✅ Improved error messages
• ✅ Updated help documentation

Support

For issues or questions:

1. Check reference documentation in references/
2. Run scripts with --help for usage
3. Review example reports in aiagvda_standard_reports/
4. Check test cases in tests/ (python3 -m pytest tests/ -v)
5. Install dependencies: pip install -r requirements.txt

*Rohoon Six Sigma - Professional Quality Management Support* *Based on AIAG-VDA, ISO, and Lean Six Sigma Standards*

📊 Figure 12-3 标准格式报告生成

功能说明

生成符合 AIAG/VDA SPC 协调标准 Figure 12-3 格式的综合分析报告，支持中英文双语。

使用方式

# 1. 生成测试数据
cd ~/.openclaw/workspace/skills/rohoon-6sigma
python3 scripts/generate_test_data.py

# 2. 批量生成报告（中英文）
python3 scripts/batch_generate_reports_v2.py

输出文件

• 英文版：tmp/SPC_Reports_EN/*.pdf
• 中文版：tmp/SPC_Reports_CN/*.pdf

报告特性

• 6 列等宽表头布局
• 字段名加粗，字段值蓝色
• 单元格自动合并
• 能力指数进度条可视化
• 字体 6 号紧凑排版

应用场景

1. 客户审核 - 提供符合 AIAG/VDA 标准的正式报告
2. 内部培训 - 展示标准报告格式要求
3. 数据分析 - 识别过程能力问题
4. 持续改进 - 跟踪改进效果