终于有人把FMEA、DFMEA和PFMEA讲清楚了!Finally, someone has clearly explained FMEA, DFMEA, and PFMEA!


AI free token/AI免费积分

感谢您的关注!

附赠【六大工具学习资料.zip】

【菜单栏 - 资料领取】备注六大工具领取

在产品开发和制造过程策划中,FMEA几乎是最常用的技术风险分析方法之一。但实际应用中,很多企业仍停留在填表、评分、应付审核的层面,这种做法偏离了FMEA的本质。

FMEA不是一张风险登记表,也不只是S、O、D评分工具,而是一套由跨职能团队实施的、系统化的定性技术风险分析方法。

DFMEA和PFMEA也并不是另外得两种独立工具,而是FMEA分别应用于产品设计和制造过程后的两种主要形式。其关系可以概括为:FMEA用于识别和降低技术风险;DFMEA关注产品设计失效;PFMEA关注制造、装配和物流过程失效。

一、FMEA的本质

FMEA是Failure Mode and Effects Analysis的缩写,即失效模式及影响分析。

它主要用于:

  • 识别产品功能或过程功能;
  • 分析潜在失效模式、失效影响和失效起因;
  • 评价现有预防控制和探测控制;
  • 确定风险降低措施及措施优先级;
  • 跟踪措施实施并重新评价风险;
  • 将风险分析结果形成技术文件。

FMEA分析的是技术风险,不直接替代财务风险、进度风险和战略风险分析。它能够支持产品安全性、可靠性、可制造性和过程稳定性的提升,但不能单独解决所有项目风险。

FMEA还具有两个重要属性。

第一,它是事前预防工具。应在设计冻结、工艺定型、设备采购和量产实施之前开展。越接近项目后期,修改设计和过程的成本越高,FMEA能够发挥的预防价值越低。

第二,它是动态技术文件。以下情形通常都需要启动或重新评审FMEA:

  1. 新设计、新技术或新过程;
  2. 现有设计或过程用于新的环境、场地或应用;
  3. 现有设计或过程发生工程变更;
  4. 顾客要求、法规、材料、设备或操作条件发生变化;
  5. 出现内部质量问题、零公里问题、客诉或使用现场失效;
  6. 获得新的试验结果或经验教训。

因此,FMEA不是项目结束后封存的表格,而应随产品和过程生命周期持续更新。

二、DFMEA

DFMEA是Design Failure Mode and Effects Analysis,即设计失效模式及影响分析。

其分析对象可以是系统、子系统、组件、零件,也可以是设备和工装。DFMEA的核心不是检查图纸有没有画错,而是分析产品设计在预期使用环境和寿命周期内,能否持续实现规定功能。

DFMEA重点关注:

  • 产品功能和技术要求是否完整;
  • 系统、子系统和零部件之间的接口是否合理;
  • 材料、结构、尺寸和公差是否满足功能;
  • 产品对载荷、温度、振动、腐蚀等条件是否具备足够适应性;
  • 设计是否对制造变差过于敏感;
  • 安全、法规、可靠性和可维护性要求是否得到落实;
  • 设计验证计划能否有效暴露潜在失效。

例如,某连接支架的设计功能是:在规定载荷和振动条件下,保持部件位置和连接稳定。

其DFMEA失效链可以描述为:

  • 失效影响FE:部件位移、松动或异响,严重时造成系统功能丧失;
  • 失效模式FM:支架变形超过允许范围、产生裂纹或发生断裂;
  • 失效起因FC:截面尺寸不足、圆角过小、材料强度或疲劳性能不足。

这里需要注意,“刚度不足”更接近设计特性不足或失效起因,不能简单与“开裂、断裂”并列为同一层级的失效模式。失效模式应描述关注要素以何种方式未能实现规定功能。

DFMEA的风险降低措施通常包括:

  • 优化结构、材料和尺寸;
  • 调整公差及接口设计;
  • 消除应力集中;
  • 增加冗余或防错设计;
  • 补充仿真、寿命和环境试验;
  • 提高设计对制造和使用变差的鲁棒性。

DFMEA应由承担设计责任的工程师或产品开发团队主导,质量、可靠性、试验、制造、采购和供应商等共同参与。

三、PFMEA

PFMEA是Process Failure Mode and Effects Analysis,即过程失效模式及影响分析。

PFMEA分析的不是设计方案是否合理,而是制造、装配、检验和内部物流过程,能否稳定地实现设计要求。

其重点风险包括:

  • 设备参数设置错误;
  • 工装定位不稳定;
  • 刀具、夹具或设备性能衰减;
  • 材料、零件或程序混用;
  • 操作步骤遗漏或顺序错误;
  • 过程防错不足;
  • 测量系统能力不足;
  • 不合格品未被及时识别和隔离;
  • 过程工作要素发生异常或变差。

AIAG—VDA PFMEA在结构分析中,通常从人员、设备、材料和环境等4M过程工作要素展开,企业也可结合实际过程扩展为5M、6M等类别。其目的不是机械套用鱼骨图,而是系统识别影响过程步骤功能的潜在起因来源。

仍以连接支架为例。设计已经规定螺栓规格和拧紧扭矩,PFMEA需要分析装配过程能否稳定满足要求。

对应的失效链可以是:

  • 失效影响FE:支架松动、异响,影响产品使用功能;
  • 失效模式FM:螺栓实际拧紧扭矩低于规定要求;
  • 失效起因FC:拧紧程序调用错误、工具未校准、套筒磨损或操作漏拧。

PFMEA的措施通常指向过程控制,例如:

  • 设置程序自动调用和参数互锁;
  • 建立工具校准与寿命管理;
  • 应用扭矩—角度监控;
  • 增加漏装、错装防错;
  • 优化工装、设备和作业顺序;
  • 明确过程监控频次和反应计划;
  • 将控制要求落实到控制计划和作业文件。

PFMEA通常由过程或制造工程师主导,质量、生产、设备、维护、设计、物流和现场人员共同参与。

四、DFMEA和PFMEA的核心区别

DFMEA和PFMEA虽然采用相同的方法框架,但分析对象、起因来源和输出方向不同。

对比维度DFMEAPFMEA
分析对象产品系统、子系统、组件或零件制造、装配、检验和物流过程
核心问题设计能否实现预期功能过程能否稳定实现设计要求
典型起因材料、结构、尺寸、公差、接口、设计原理人员、设备、材料、环境及其他过程工作要素
主导责任设计或研发团队工艺或制造团队
主要输入顾客要求、技术规范、边界图、P图、试验要求过程流程图、图纸、工艺方案、设备和工装规划
主要输出设计变更、规范修订、验证计划防错、参数控制、控制计划、作业和检验标准

最简洁的判断方式是:DFMEA研究产品是否被正确设计,PFMEA研究产品能否被过程正确制造出来。

PFMEA不能用于掩盖设计缺陷。

产品结构强度不足导致断裂,本质上属于设计风险,增加终检不能从根本上消除;设计要求合理,但生产过程中发生错料、错装或参数失控,则属于过程风险,也不能简单归因于设计。

五、失效影响、失效模式和失效起因必须形成失效链

FMEA中最常见的专业错误,是把FE、FM和FC写成同一层级的现象。

三者分别回答:

  • 失效影响FE:该失效对上一级系统、后续过程或最终用户造成什么后果?
  • 失效模式FM:关注要素以什么方式未能实现功能?
  • 失效起因FC:什么原因导致关注要素发生失效?

其因果关系为:失效起因FC → 失效模式FM → 失效影响FE

例如,在拧紧工序中:

  • 程序调用错误,是失效起因;
  • 螺栓扭矩不足,是失效模式;
  • 连接松动并导致异响,是失效影响。

同一个现象在不同分析层级下,可能承担不同角色。“连接松动”在零部件层级可能是失效模式,在整车系统层级可能是导致其他功能异常的起因,对某一道装配工序而言又可能是失效影响。

因此,判断FE、FM和FC之前,必须先明确:

  1. 当前分析的关注要素是什么;
  2. 上一较高级别要素是什么;
  3. 下一较低级别要素或过程工作要素是什么;
  4. 当前分析边界包括什么、不包括什么。

没有结构层级,就无法建立准确的失效链。

六、FMEA七步法

AIAG—VDA FMEA方法采用七步法,将DFMEA、PFMEA和FMEA-MSR纳入统一的分析框架。

第一步:规划和准备通过5T明确FMEA的目的、时间安排、团队、任务和工具,同时确定分析范围、项目边界和不纳入分析的内容。

第二步:结构分析DFMEA建立系统、系统要素和组件要素之间的层级关系;PFMEA建立过程项、过程步骤和过程工作要素之间的关系。

第三步:功能分析定义各层级产品要素或过程步骤的功能和要求。没有功能定义,就无法准确判断功能如何失效。

第四步:失效分析围绕功能建立FE—FM—FC失效链,分析失效如何从起因传递到上一层级、后续过程或最终用户。

第五步:风险分析识别当前预防控制和探测控制,并进行S、O、D评级:

  • 严重度S:评价失效影响FE的严重程度;
  • 频度O:评价失效起因FC的发生频度,并考虑现有预防控制;
  • 探测度D:评价当前探测控制发现失效起因FC或失效模式FM的能力。

不能把S、O、D笼统地理解为对整行失效模式统一评分。

AIAG—VDA方法以措施优先级AP作为主要措施排序逻辑。AP综合考虑S、O和D,但不是三者的简单乘积。若顾客特殊要求或企业程序仍要求计算RPN,可以作为补充信息保留,但不宜仅按照RPN大小决定是否采取措施。

第六步:优化针对需要进一步降低的风险,制定预防措施和探测措施,明确责任人、完成时间、状态和验证证据。

通常情况下:

  • 改变失效影响或产品设计,才可能降低S;
  • 消除失效起因、强化预防控制或实施防错,主要降低O;
  • 增加有效检测、监控和验证机会,主要改善D。

措施完成后不能为了获得更低AP而主观下调评分,必须依据设计、预防控制或探测控制是否发生实质变化重新评价。

第七步:结果文件化总结分析范围、主要风险、未关闭措施、风险降低结果和残余风险,并向管理层、顾客或相关方沟通。

七步法的技术逻辑是:先分析结构和功能,再建立失效链;先评价风险,再实施并验证措施。

七、DFMEA和PFMEA必须建立信息传递关系

DFMEA和PFMEA不是两份相互独立的文件,而是产品设计风险向过程控制传递的两个环节。

典型的信息链为:顾客要求 → 产品功能 → DFMEA → 设计要求与产品特性 → PFMEA → 控制计划 → 作业指导书与检验标准

DFMEA识别出的产品功能、失效影响、设计要求以及对制造变差敏感的特性,应传递到PFMEA。

PFMEA继续分析:

  • 哪些过程步骤形成该产品特性;
  • 哪些过程工作要素可能造成特性失效;
  • 应采用什么预防和探测控制;
  • 发生异常后如何反应和处置。

当DFMEA与PFMEA描述的是同一个最终用户失效影响时,严重度评价原则上应保持一致。严重度评价的是失效影响,而不是失效发生概率;频度和探测度则应依据各自的设计控制或过程控制分别评价。

对于需要特别过程控制的产品或过程特性,还应在PFMEA、控制计划和相关现场文件中保持一致标识,并落实监控、记录和反应计划。

如果DFMEA中的高风险内容没有进入PFMEA,或者PFMEA中的控制要求没有进入控制计划和现场文件,风险传递链就已经中断。

八、为什么很多企业的FMEA没有效果?

FMEA失效的根本原因通常不在评分表,而在管理方式。

1. 由质量部门单独编制:质量人员可以组织和推进FMEA,但不能替代设计、工艺、设备、试验和现场人员完成技术分析。

2. 开展时间过晚:设计冻结、设备采购或量产之后再做FMEA,往往只能增加检验和筛选,难以从源头降低风险。

3. 直接复制历史模板:基础FMEA和家族FMEA可以复用,但必须识别产品、环境、材料、设备和工艺差异。未经差异分析的复制,会把旧项目的假设和漏洞一起带入新项目。

4. 只依靠增加检验降低风险:检验属于探测控制,主要作用是发现问题,不能消除失效起因。风险降低应优先考虑设计优化、起因消除、预防控制和自动防错。

5. 过度关注评分和RPN:评分只是辅助决策工具。FMEA的真正输出应是设计优化、过程防错、验证方案和控制要求,而不是一个更低的数字。

6. FMEA与现场文件脱节:设计、材料、设备、工艺或客诉发生变化后,PFMEA、控制计划和作业文件没有同步更新,FMEA就会迅速失去有效性。

7. 管理责任不清:FMEA团队负责技术分析,产品或过程所有者负责措施落实,管理层负责资源配置以及残余风险接受。重大风险不能仅由质量工程师或表格编制人员签字关闭。

写在最后

FMEA、DFMEA和PFMEA之间的关系并不复杂:FMEA是技术风险分析方法,DFMEA用于降低产品设计风险,PFMEA用于降低制造和装配过程风险。

DFMEA回答产品功能为什么可能失效,PFMEA回答制造过程为什么可能无法稳定实现设计要求。两者通过产品功能、失效影响、设计要求和产品特性建立联系,并最终落实到设计验证、过程防错、控制计划和现场作业标准。

一份高质量FMEA的价值,不在于表格有多少行,也不在于评分有多低,而在于是否真正推动了设计优化、试验验证、过程控制、防错措施和经验沉淀。

推荐阅读:

1.形位公差及检测:从图样标注、公差带分析到现场测量,一篇讲清

2.APQP五阶段详解:一篇讲透每个阶段做什么、输出什么、评审什么

3.计量型MSA和计数型MSA到底有什么区别?一篇讲清

4.设计验证DV和生产验证PV是什么?两者有什么区别?

5.Cgk、Cmk、Ppk、Cpk到底有什么区别?能力指数一次讲清楚

6.终于有人把质量策划、质量计划和控制计划讲清楚了

Thank you for your attention!

Bonus: 【Six Essential Tools Study Materials.zip】

【Menu Bar - Materials Access】Mention "Six Essential Tools" to receive them.

In product development and manufacturing process planning, FMEA is one of the most commonly used technical risk analysis methods. However, in practical application, many companies still remain at the level of filling out forms, scoring, and coping with audits, which deviates from the essence of FMEA.

FMEA is not a risk register, nor is it just an S, O, D scoring tool. It is a systematic, qualitative technical risk analysis method implemented by a cross-functional team.

DFMEA and PFMEA are not two additional independent tools either. They are the two main forms of FMEA applied to product design and manufacturing processes respectively. The relationship can be summarized as follows: FMEA is used to identify and reduce technical risks; DFMEA focuses on product design failures; PFMEA focuses on manufacturing, assembly, and logistics process failures.

1. The essence of FMEA

FMEA is the abbreviation of Failure Mode and Effects Analysis.

It is mainly used to:

Identify product functions or process functions;

Analyze potential failure modes, failure effects, and failure causes;

Evaluate existing prevention controls and detection controls;

Determine risk reduction measures and measure priorities;

Track the implementation of measures and re-evaluate risks;

Document the risk analysis results as technical documentation.

FMEA analyzes technical risks. It does not directly replace financial risk, schedule risk, or strategic risk analysis. It can support improvements in product safety, reliability, manufacturability, and process stability, but it cannot alone solve all project risks.

FMEA also has two important attributes.

First, it is a proactive prevention tool. It should be carried out before design freeze, process finalization, equipment procurement, and mass production implementation. The closer to the later stages of a project, the higher the cost of modifying the design and process, and the lower the preventive value FMEA can provide.

Second, it is a dynamic technical document. FMEA usually needs to be initiated or re-reviewed under the following circumstances:

New design, new technology, or new process;

Existing design or process is used in a new environment, location, or application;

Engineering changes occur to the existing design or process;

Customer requirements, regulations, materials, equipment, or operating conditions change;

Internal quality issues, zero-kilometer issues, customer complaints, or field failures occur;

New test results or lessons learned are obtained.

Therefore, FMEA is not a form to be archived after the project ends, but should be continuously updated throughout the product and process life cycle.

2. DFMEA

DFMEA is Design Failure Mode and Effects Analysis.

Its analysis objects can be systems, subsystems, components, parts, as well as equipment and tooling. The core of DFMEA is not to check whether drawings are drawn incorrectly, but to analyze whether the product design can continuously achieve the specified functions within the intended use environment and life cycle.

DFMEA focuses on:

Whether product functions and technical requirements are complete;

Whether the interfaces among systems, subsystems, and components are reasonable;

Whether materials, structures, dimensions, and tolerances meet the functions;

Whether the product has sufficient adaptability to loads, temperature, vibration, corrosion, and other conditions;

Whether the design is overly sensitive to manufacturing variation;

Whether safety, regulatory, reliability, and maintainability requirements are implemented;

Whether the design verification plan can effectively expose potential failures.

For example, the designed function of a connecting bracket is to maintain component position and connection stability under specified load and vibration conditions.

Its DFMEA failure chain can be described as:

Failure Effect (FE): Component displacement, loosening, or abnormal noise; in severe cases, loss of system function;

Failure Mode (FM): Bracket deformation exceeds the allowable range, cracks develop, or fracture occurs;

Failure Cause (FC): Insufficient cross-sectional dimensions, excessively small fillet radius, or insufficient material strength or fatigue performance.

Here, it should be noted that "insufficient stiffness" is closer to a deficiency in design characteristics or a failure cause, and it cannot simply be listed at the same level as failure modes such as "cracking" or "fracture." The failure mode should describe the manner in which the element of focus fails to achieve the specified function.

Risk reduction measures for DFMEA usually include:

Optimizing structure, materials, and dimensions;

Adjusting tolerances and interface design;

Eliminating stress concentrations;

Adding redundancy or mistake-proofing design;

Supplementing simulation, life, and environmental tests;

Improving the robustness of the design to manufacturing and usage variation.

DFMEA should be led by the engineer or product development team responsible for design, with quality, reliability, testing, manufacturing, purchasing, and suppliers participating together.

3. PFMEA

PFMEA is Process Failure Mode and Effects Analysis.

PFMEA does not analyze whether the design solution is reasonable, but whether manufacturing, assembly, inspection, and internal logistics processes can stably achieve the design requirements.

Its key risks include:

Incorrect equipment parameter settings;

Unstable tooling positioning;

Performance degradation of cutting tools, fixtures, or equipment;

Mixing up materials, parts, or programs;

Omitted operation steps or incorrect sequence;

Insufficient process mistake-proofing;

Insufficient measurement system capability;

Nonconforming products not identified and isolated in a timely manner;

Abnormalities or variation in process work elements.

In structural analysis, AIAG-VDA PFMEA usually starts from the 4M process work elements of personnel, equipment, materials, and environment. Companies can also expand to 5M, 6M, or other categories based on their actual processes. The purpose is not to mechanically apply fishbone diagrams, but to systematically identify potential cause sources that affect the function of process steps.

Still using the connecting bracket as an example. The design has already specified the bolt specification and tightening torque, and PFMEA needs to analyze whether the assembly process can stably meet the requirements.

The corresponding failure chain can be:

Failure Effect (FE): Bracket loosening or abnormal noise, affecting product usage function;

Failure Mode (FM): The actual tightening torque of the bolt is below the specified requirement;

Failure Cause (FC): Incorrect tightening program call, uncalibrated tool, socket wear, or omitted tightening operation.

PFMEA measures usually point to process control, such as:

Setting up automatic program call and parameter interlocking;

Establishing tool calibration and life management;

Applying torque-angle monitoring;

Adding mistake-proofing for missing or incorrect installation;

Optimizing tooling, equipment, and operation sequence;

Clarifying process monitoring frequency and reaction plans;

Implementing control requirements into control plans and work instructions.

PFMEA is usually led by process or manufacturing engineers, with quality, production, equipment, maintenance, design, logistics, and shop-floor personnel participating together.

4. Core differences between DFMEA and PFMEA

Although DFMEA and PFMEA use the same method framework, their analysis objects, cause sources, and output directions are different.

The simplest way to distinguish them is: DFMEA studies whether the product is correctly designed, while PFMEA studies whether the product can be correctly manufactured by the process.

PFMEA cannot be used to cover up design deficiencies.

Fracture caused by insufficient structural strength of the product is essentially a design risk, and adding final inspection cannot fundamentally eliminate it. If the design requirements are reasonable, but wrong materials, incorrect assembly, or parameter loss of control occur during production, that is a process risk and cannot simply be attributed to design.

5. Failure effects, failure modes, and failure causes must form a failure chain

The most common professional error in FMEA is writing FE, FM, and FC as phenomena at the same level.

The three answer respectively:

Failure Effect (FE): What consequences does this failure have on the higher-level system, subsequent process, or end user?

Failure Mode (FM): In what manner does the element of focus fail to achieve its function?

Failure Cause (FC): What cause leads to the failure of the element of focus?

The causal relationship is: Failure Cause (FC) → Failure Mode (FM) → Failure Effect (FE)

For example, in a tightening operation:

An incorrect program call is the failure cause;

Insufficient bolt torque is the failure mode;

Connection loosening leading to abnormal noise is the failure effect.

The same phenomenon can play different roles at different analysis levels. "Connection loosening" may be a failure mode at the component level, a cause of other functional abnormalities at the vehicle system level, and a failure effect for a particular assembly operation.

Therefore, before determining FE, FM, and FC, the following must first be clarified:

What is the element of focus in the current analysis;

What is the next higher-level element;

What is the next lower-level element or process work element;

What is included in the current analysis boundary and what is not;

Without structural hierarchy, an accurate failure chain cannot be established.

6. The seven-step FMEA method

The AIAG-VDA FMEA method uses a seven-step approach, bringing DFMEA, PFMEA, and FMEA-MSR into a unified analysis framework.

Step 1: Planning and Preparation. Use the 5Ts to clarify the purpose, timing, team, tasks, and tools of the FMEA, while determining the analysis scope, project boundaries, and content excluded from the analysis.

Step 2: Structural Analysis. DFMEA establishes the hierarchical relationships among the system, system elements, and component elements; PFMEA establishes the relationships among process items, process steps, and process work elements.

Step 3: Function Analysis. Define the functions and requirements of each product element or process step at each level. Without function definitions, it is impossible to accurately determine how functions fail.

Step 4: Failure Analysis. Build the FE-FM-FC failure chain around the functions, and analyze how failures propagate from causes to the higher level, subsequent process, or end user.

Step 5: Risk Analysis. Identify current prevention controls and detection controls, and perform S, O, D ratings:

Severity (S): Evaluate the severity of the failure effect FE;

Occurrence (O): Evaluate the frequency of the failure cause FC, considering existing prevention controls;

Detection (D): Evaluate the ability of current detection controls to detect the failure cause FC or failure mode FM.

S, O, and D cannot be broadly understood as a unified score for an entire row of failure modes.

The AIAG-VDA method uses Action Priority (AP) as the primary logic for prioritizing measures. AP considers S, O, and D comprehensively, but it is not a simple product of the three. If customer-specific requirements or company procedures still require calculating RPN, it may be retained as supplementary information, but decisions on whether to take action should not be based solely on the RPN value.

Step 6: Optimization. For risks that need further reduction, develop prevention measures and detection measures, and clarify the responsible person, completion date, status, and verification evidence.

Generally:

Only changing the failure effect or the product design can reduce S;

Eliminating failure causes, strengthening prevention controls, or implementing mistake-proofing mainly reduces O;

Adding effective inspection, monitoring, and verification opportunities mainly improves D.

After measures are completed, scores must not be subjectively lowered just to obtain a lower AP. Re-evaluation must be based on whether the design, prevention controls, or detection controls have substantially changed.

Step 7: Documentation of Results. Summarize the analysis scope, key risks, open measures, risk reduction results, and residual risks, and communicate them to management, customers, or relevant stakeholders.

The technical logic of the seven-step method is: first analyze structure and function, then establish the failure chain; first evaluate risk, then implement and verify measures.

7. DFMEA and PFMEA must establish an information transfer relationship

DFMEA and PFMEA are not two independent documents, but two links in the transfer of product design risk into process control.

The typical information chain is: Customer Requirements → Product Functions → DFMEA → Design Requirements and Product Characteristics → PFMEA → Control Plan → Work Instructions and Inspection Standards

The product functions, failure effects, design requirements, and characteristics sensitive to manufacturing variation identified by DFMEA should be transferred to PFMEA.

PFMEA then analyzes:

Which process steps create that product characteristic;

Which process work elements could cause characteristic failure;

What prevention and detection controls should be used;

How to react and dispose of abnormalities after they occur.

When DFMEA and PFMEA describe the same end-user failure effect, the severity ratings should in principle remain consistent. Severity evaluates the failure effect, not the probability of failure occurrence; frequency and detection should be evaluated separately according to their respective design controls or process controls.

For product or process characteristics requiring special process control, consistent identification should also be maintained in the PFMEA, control plan, and related shop-floor documents, and monitoring, recording, and reaction plans should be implemented.

If high-risk content from DFMEA does not enter PFMEA, or if control requirements from PFMEA do not enter the control plan and shop-floor documents, the risk transfer chain is already broken.

8. Why does FMEA have no effect in many companies?

The root cause of FMEA failure is usually not the scoring table, but the management approach.

1. Prepared solely by the quality department: Quality personnel can organize and drive FMEA, but they cannot replace design, process, equipment, testing, and shop-floor personnel in completing technical analysis.

2. Conducted too late: Performing FMEA after design freeze, equipment procurement, or mass production often only adds inspection and screening, making it difficult to reduce risk at the source.

3. Directly copying historical templates: Basic FMEA and family FMEA can be reused, but differences in product, environment, materials, equipment, and process must be identified. Copying without difference analysis brings the assumptions and gaps of old projects into new projects.

4. Relying only on adding inspections to reduce risk: Inspection is a detection control. Its main role is to find problems, not to eliminate failure causes. Risk reduction should prioritize design optimization, cause elimination, prevention controls, and automatic mistake-proofing.

5. Overemphasizing scores and RPN: Scoring is only a decision-support tool. The true output of FMEA should be design optimization, process mistake-proofing, verification plans, and control requirements, not a lower number.

6. FMEA disconnected from shop-floor documents: When design, materials, equipment, process, or customer complaints change, if the PFMEA, control plan, and work instructions are not updated simultaneously, FMEA quickly loses its effectiveness.

7. Unclear management responsibilities: The FMEA team is responsible for technical analysis, product or process owners are responsible for measure implementation, and management is responsible for resource allocation and acceptance of residual risk. Major risks cannot be closed by the signature of a quality engineer or form compiler alone.

Final remarks

The relationship among FMEA, DFMEA, and PFMEA is not complex: FMEA is a technical risk analysis method, DFMEA is used to reduce product design risk, and PFMEA is used to reduce manufacturing and assembly process risk.

DFMEA answers why product functions may fail, and PFMEA answers why the manufacturing process may be unable to stably achieve design requirements. The two are connected through product functions, failure effects, design requirements, and product characteristics, and are ultimately implemented into design verification, process mistake-proofing, control plans, and shop-floor work standards.

The value of a high-quality FMEA is not in how many rows the table has, nor in how low the scores are, but in whether it truly drives design optimization, test verification, process control, mistake-proofing measures, and experience accumulation.

Recommended reading:

1. GD&T and inspection: from drawing callouts, tolerance zone analysis to shop-floor measurement, explained clearly in one article

2. Detailed explanation of the five APQP phases: what is done, what is output, and what is reviewed in each phase, explained thoroughly in one article

3. What exactly is the difference between variable MSA and attribute MSA? Explained clearly in one article

4. What are Design Verification (DV) and Production Verification (PV)? What is the difference between them?

5. What exactly is the difference between Cgk, Cmk, Ppk, and Cpk? Capability indices explained clearly once and for all

6. Finally, someone has explained quality planning, quality plans, and control plans clearly

暂无评论,快来发表第一条评论吧!No comments yet. Be the first!

📮 需求咨询📮 Inquiry