InfiniSynapse
What is a 5 Whys template?什么是 5 Whys Template 五问法模板？
For the full topic map and the neighboring methods that support this workflow, continue with the anomaly detection and root cause analysis guide.
如需查看完整主题结构以及支撑本流程的相邻方法，请继续阅读异常检测与根因分析指南。
A 5 Whys template is a repeatable root cause analysis worksheet that records a factual problem, a linked sequence of “why” questions and evidence, a supported process cause, corrective action, and an effectiveness check. The number five is a prompt, not a rule: stop when the team reaches a cause it can support and control, or branch and continue when the evidence reveals more than one path.
5 Whys Template 五问法模板是一份可重复使用的根因分析工作表，用于记录事实性问题、相互连接的“为什么”问题与证据、得到支持的流程原因、纠正措施和有效性检查。数字“五”只是提示而不是规则：当团队到达一个有证据支持且可控制的原因时可以停止；如果证据显示存在多条路径，就应分支并继续追问。
A useful template needs more than five blank boxes. A strong five whys analysis template preserves the logic between answers, identifies what evidence would confirm or reject each link, separates contributing factors from root causes, and connects the finding to an owner, due date, baseline, success measure, and review date. Without those fields, the exercise can become a polished chain of opinions.
实用模板不能只有五个空白框。可靠的 five whys analysis template 应保留答案之间的逻辑，标明什么证据能够确认或否定每个连接，区分促成因素与根因，并把结论连接到负责人、截止日期、基线、成功指标和复核日期。缺少这些字段，练习很容易变成一条看似整齐的主观看法链。
Copy this evidence-ready 5 Whys template复制这份支持证据记录的五问法模板
Begin with one event or outcome, not a proposed cause. Paste the structure into a document, spreadsheet, incident record, or workshop board. If the chain branches, duplicate the why rows for each supported path rather than forcing every observation into one story.
从一个事件或结果开始，而不是从预设原因开始。可把下面结构复制到文档、电子表格、事件记录或研讨白板中。如果因果链发生分支，应为每条得到支持的路径复制“为什么”行，不要把所有观察强行塞入同一个故事。
| Field字段 | Record记录内容 | Evidence check证据检查 |
|---|---|---|
| Problem and scope问题与范围 | What happened, where, when, how often, and to what measurable extent?发生了什么、地点、时间、频率和可衡量程度是什么？ | Baseline, timestamps, records, affected population基线、时间戳、记录、受影响对象 |
| Why 1 | Why did the observable problem occur?为什么会出现可观察问题？ | Fact supporting the answer; plausible alternative支持答案的事实；合理替代解释 |
| Why 2 | Why did the condition in answer 1 exist?为什么答案 1 中的条件会存在？ | Source, owner, date, confidence来源、责任人、日期、置信程度 |
| Why 3 | What process or control allowed answer 2?什么流程或控制允许答案 2 发生？ | Process record, comparison, or test流程记录、对照或检验 |
| Why 4 | Why was that process or control designed or operated this way?为什么该流程或控制以这种方式设计或运行？ | Policy, configuration, handoff, incentive政策、配置、交接、激励因素 |
| Why 5 | What supported system cause can the organization change?组织可以改变的、得到支持的系统原因是什么？ | Mechanism, counterfactual, alternatives作用机制、反事实、替代解释 |
| Action and verification措施与验证 | Countermeasure, owner, due date, review date对策、负责人、截止日期、复核日期 | Baseline, target, observation window, side effects基线、目标、观察窗口、副作用 |
Stopping test: the final answer should describe a specific process, control, resource, design, or information condition that the organization can influence. “Human error,” “carelessness,” and “lack of attention” usually stop too early unless evidence identifies the system conditions that shaped the behavior.
停止检验：最终答案应描述组织可以影响的具体流程、控制、资源、设计或信息条件。“人为错误”“粗心”和“注意力不足”通常意味着停止过早，除非证据已经识别出塑造该行为的系统条件。
When to use a 5 Whys root cause analysis template何时使用五问法根因分析模板
Use the method after a bounded problem is observed and enough facts exist to test a causal chain. It works best for a relatively understandable process, one primary outcome, and a team with direct process knowledge. Prepare the incident timeline, relevant measurements, system or process records, recent changes, current procedures, and people who saw the work as performed—not only managers who know the written procedure.
当一个边界清晰的问题已经被观察到，而且有足够事实可用于检验因果链时，可以使用该方法。它最适合相对可理解的流程、一个主要结果，以及拥有直接流程知识的团队。开始前应准备事件时间线、相关测量、系统或流程记录、近期变更、现行程序，以及真正观察过实际工作的人，而不只是了解书面流程的管理者。
A recurring report delay, a missed handoff, a known configuration failure, a production defect with a bounded path, or a service breakdown with a reliable timeline.
重复出现的报告延迟、交接遗漏、已知配置故障、路径边界清晰的生产缺陷，或具有可靠时间线的服务中断。
Use a fishbone diagram for many plausible branches, fault tree analysis for logical combinations, or a formal investigation for high-harm, regulated, contested, or technically complex events.
当存在多个合理分支时先用鱼骨图；涉及逻辑组合时用故障树；对于高危害、受监管、有争议或技术复杂的事件，应开展正式调查。
Do not use the template to assign blame, retrofit a preferred solution, or replace required safety, legal, clinical, security, or regulatory investigations. For a high-stakes event, preserve evidence and follow the applicable incident process before facilitating an informal workshop.
不要用模板追责个人、为偏好的方案倒推原因，也不要用它替代必要的安全、法律、临床、安全防护或监管调查。对于高风险事件，应先保存证据并遵循适用的事件处理流程，再考虑开展非正式研讨。
How to complete the 5 Whys template step by step如何逐步填写 5 Whys Template
- Write a neutral problem statement.写出中性问题陈述。State the observed gap with time, place, process, and magnitude. Avoid causes and solutions. “Daily reconciliation completed 95 minutes late on August 7” is testable; “the team needs better training” is already a conclusion.用时间、地点、流程和程度描述观察到的差距，避免写入原因和方案。“8 月 7 日每日对账晚了 95 分钟”可检验；“团队需要更好培训”已经是结论。
- Ask why and record one precise answer.追问原因并记录一个精确答案。Base the answer on what happened in this event. Mark an unverified claim as a hypothesis. If two conditions independently mattered, branch the chain instead of joining them with a vague “and.”答案应基于本次事件实际发生的情况。未经验证的说法要标记为假设。如果两个条件独立产生影响，应把因果链分支，而不是用模糊的“以及”连接。
- Check the causal link before asking again.继续追问前先检查因果连接。Ask what record, observation, comparison, or experiment supports the claim. Also ask what else could produce the same pattern. A sequence is only as reliable as its weakest link.询问什么记录、观察、对照或实验支持该说法，同时询问还有什么可能产生相同模式。整条链的可靠性取决于其中最薄弱的连接。
- Continue until the mechanism is actionable.继续追问直到机制可行动。Move from symptom to local condition, then to process or system conditions. You may stop before five or continue beyond five. The endpoint should explain why the event could occur and identify a control the organization can change.从症状进入局部条件，再进入流程或系统条件。可以少于五次，也可以超过五次。终点应解释事件为何能够发生，并识别组织可以改变的控制措施。
- Design and verify a countermeasure.设计并验证对策。Choose an action linked to the supported mechanism, name an owner and date, define a baseline and target, watch for recurrence and side effects, and schedule a decision to adopt, adjust, or reverse the change.选择与得到支持的机制相连的措施，指定负责人和日期，定义基线与目标，监测复发和副作用，并安排一次复核，以决定采纳、调整或撤销变更。
5 Whys template example: a delayed daily report五问法模板示例：每日报告延迟
The following is a hypothetical training example, not a customer case. Assume a daily operations report was delivered 95 minutes late on August 7. The team has job logs, file-arrival timestamps, a scheduler configuration history, and the approved data-source inventory.
以下是假设性的培训示例，不是客户案例。假设 8 月 7 日的每日运营报告延迟 95 分钟交付。团队拥有作业日志、文件到达时间戳、调度器配置历史和已批准的数据源清单。
| Question问题 | Answer答案 | Evidence in this example本示例证据 |
|---|---|---|
| Why 1 | The report job waited for a required source file.报告作业等待一个必需的源文件。 | Job log shows the file dependency blocked execution.作业日志显示文件依赖阻塞执行。 |
| Why 2 | The file arrived after the job cutoff.文件在作业截止时间之后到达。 | Arrival timestamp is later than the configured cutoff.到达时间戳晚于配置的截止时间。 |
| Why 3 | The upstream schedule had changed, but this dependency had not.上游调度已经变更，但该依赖没有同步变更。 | Configuration history shows only the upstream change.配置历史只显示上游变更。 |
| Why 4 | The change review did not list downstream report dependencies.变更评审没有列出下游报告依赖。 | Review record lacks a dependency-impact field.评审记录缺少依赖影响字段。 |
| Why 5 | No maintained owner-and-dependency register fed the review process.没有维护责任人与依赖登记表供评审流程使用。 | Inventory lists sources, but not downstream owners or schedules.清单列出数据源，却没有下游责任人或调度信息。 |
A proposed countermeasure is to add downstream owner and schedule fields to the inventory, require an automated impact check in schedule-change reviews, and alert both owners when a cutoff conflict appears. Example verification: compare on-time completion before and after the change over a predefined observation window, while also checking false alerts and review time. These numbers and dates must be set from real baseline data; they are not supplied by the template.
建议的对策是在清单中增加下游责任人和调度字段，在调度变更评审中要求自动影响检查，并在出现截止时间冲突时提醒双方责任人。示例验证方式是在预先定义的观察窗口内比较变更前后的按时完成情况，同时检查误报和评审耗时。具体数字和日期必须依据真实基线设定，模板不会代替团队提供这些数据。
5 Whys vs fishbone, fault tree, and full RCA五问法与鱼骨图、故障树和完整 RCA 的比较
| Method方法 | Best use最佳用途 | Main limitation主要局限 |
|---|---|---|
| 5 Whys | A bounded problem and a traceable causal path边界清晰的问题和可追踪因果路径 | Can oversimplify branches and depend on facilitator bias可能过度简化分支并受主持人偏差影响 |
| Fishbone diagram鱼骨图 | Broad exploration across several cause categories跨多个原因类别进行广泛探索 | Maps hypotheses but does not verify causality只能映射假设，不能验证因果 |
| Fault tree analysis故障树分析 | Top-down logic with AND/OR combinations包含 AND/OR 组合的自上而下逻辑 | Needs careful system modeling and reliable inputs需要严谨系统建模和可靠输入 |
| Formal RCA正式根因分析 | High-harm, regulated, cross-system, or contested events高危害、受监管、跨系统或有争议事件 | More time, governance, and specialist evidence需要更多时间、治理和专业证据 |
The methods are complementary. A team can use a timeline to establish facts, a fishbone to generate branches, 5 Whys to deepen supported branches, and a test plan to verify the mechanism. Method names do not create evidence; the quality of records, comparisons, and follow-up determines whether a conclusion is useful.
这些方法可以互补。团队可先用时间线建立事实，用鱼骨图生成分支，再用五问法深入得到支持的分支，最后通过检验计划验证机制。方法名称本身不会创造证据；记录、对照和后续复核的质量才决定结论是否有用。
Common 5 Whys mistakes and safeguards五问法常见错误与防护措施
- Starting with a cause: rewrite the opening statement as an observable gap with scope and magnitude.从原因开始：把开场陈述改写为包含范围与程度的可观察差距。
- Following the loudest opinion: record evidence and alternatives for every link; invite people with direct process knowledge.追随最响亮的意见：为每个连接记录证据和替代解释，并邀请拥有直接流程知识的人。
- Forcing a single chain: branch when independent conditions contribute. A neat worksheet is not worth losing causal structure.强行保留单一因果链：当独立条件共同促成结果时进行分支，不要为了表格整齐而丢失因果结构。
- Stopping at a person: ask what workload, interface, procedure, information, control, or incentive made the action likely.停在个人身上：继续询问什么工作负荷、界面、程序、信息、控制或激励使该行为更可能发生。
- Calling an action complete without verification: define recurrence and side-effect measures before implementation, then review them after a suitable observation window.措施完成却不验证：实施前定义复发与副作用指标，在合适观察窗口后进行复核。
Facilitation safeguard: let participants challenge a link without challenging a person. Record “unknown” when evidence is missing. A visible gap is safer than an invented answer, because it becomes a specific data-collection task.
主持防护：允许参与者质疑因果连接，但不要针对个人。缺少证据时记录“未知”。可见的空白比虚构答案更安全，因为它能转化为明确的数据收集任务。
How to verify a 5 Whys root-cause conclusion如何验证五问法的根因结论
Review the completed chain in both directions. Forward, each answer should make the next event more likely through a plausible mechanism. Backward, changing the final condition should predict a measurable difference upstream. Check whether the chain explains timing, location, affected cases, and unaffected comparisons. Record disconfirming evidence rather than hiding it.
应从两个方向复核完成的因果链。向前看，每个答案应通过合理机制使下一事件更可能发生；向后看，改变最终条件应能预测上游出现可衡量差异。检查因果链是否解释了时间、地点、受影响案例和未受影响对照，并记录否定性证据，不要隐藏它。
Before implementing a corrective action, document the baseline, measurement definition, expected direction and size of change, owner, implementation date, observation window, decision threshold, and possible side effects. After implementation, compare like with like and account for concurrent changes. Improvement supports the mechanism, but a single quiet period does not prove permanent prevention. Reopen the analysis if recurrence persists or a side effect outweighs the benefit.
实施纠正措施前，应记录基线、测量定义、预期变化方向与幅度、负责人、实施日期、观察窗口、决策阈值和潜在副作用。实施后应进行同类比较，并考虑同期发生的其他变化。改善可以支持机制，但一次平静期不能证明问题永久消失。如果仍然复发，或副作用大于收益，应重新开启分析。
Analyze the evidence behind your 5 Whys chain分析五问法因果链背后的证据
Prepare the completed worksheet, incident timeline, timestamps, process records, relevant files, and a clear question. InfiniSynapse's online analysis workspace can support analysis across connected databases, files, documents, audio, and video; it does not replace causal judgment or required investigations. Use it to examine the available evidence and comparisons that support or challenge each link.
请先准备填写完成的工作表、事件时间线、时间戳、流程记录、相关文件和清晰问题。InfiniSynapse 在线分析工作区支持分析已连接的数据库、文件、文档、音频和视频；它不能替代因果判断或必要调查。可用它检查支持或质疑每个连接的现有证据与对照。
Open the InfiniSynapse analysis workspace打开 InfiniSynapse 分析工作区5 Whys template FAQ5 Whys Template 常见问题
What is a 5 Whys template?什么是五问法模板？
It is a structured worksheet that starts with a factual problem statement and records linked why questions, evidence, a supported root-cause statement, corrective action, and effectiveness checks.
它是一份结构化工作表，从事实性问题陈述开始，记录相互连接的“为什么”问题、证据、得到支持的根因陈述、纠正措施和有效性检查。
Do I have to ask exactly five whys?必须恰好追问五次吗？
No. Five is a practical label, not a stopping rule. Stop at a supported, controllable process cause; continue or branch when evidence shows more depth or multiple causal paths.
不必。五只是实用标签，不是停止规则。到达有证据支持且可控制的流程原因时可以停止；当证据显示更深层原因或多条因果路径时，应继续或分支。
What should a 5 Whys template include?五问法模板应包含哪些字段？
Include event and scope, baseline evidence, each question and answer, evidence for every link, assumptions and alternatives, supported cause, action, owner, due date, success measure, observation window, and review decision.
应包括事件与范围、基线证据、每个问题和答案、每个连接的证据、假设与替代解释、得到支持的原因、措施、负责人、截止日期、成功指标、观察窗口和复核决定。
When should I use a fishbone diagram instead?何时应改用鱼骨图？
Use a fishbone first when several independent cause categories are plausible or the team needs broad exploration. Then apply 5 Whys to supported branches and test the resulting claims.
当多个独立原因类别都合理，或团队需要广泛探索时，先使用鱼骨图。随后对得到支持的分支应用五问法，并检验形成的因果主张。
How do I know the root cause is verified?如何判断根因已经得到验证？
Confidence improves when evidence confirms each link, alternatives have been considered, the mechanism predicts the observed pattern, and a controlled action reduces recurrence without unacceptable side effects.
当证据确认每个连接、替代解释已被考虑、机制能够预测观察模式，而且受控措施在没有不可接受副作用的情况下减少复发时，结论可信度会提高。
Official sources and next steps权威来源与下一步
- U.S. Centers for Medicare & Medicaid Services: Five Whys tool — provides a printable worksheet, worked example, and a root-cause validation question.美国医疗保险和医疗补助服务中心：Five Whys 工具——提供可打印工作表、完整示例和根因验证问题。
- U.S. Centers for Medicare & Medicaid Services: Guidance for Performing RCA — describes fact gathering, contributing factors, system causes, corrective changes, and effectiveness measurement.美国医疗保险和医疗补助服务中心：RCA 实施指南——介绍事实收集、促成因素、系统原因、纠正变更和有效性测量。
Use the template as a decision record, not a ritual. Preserve uncertainty, test the links that matter, assign ownership, and judge the result by verified change rather than by whether every box is filled.
应把模板作为决策记录，而不是形式化仪式。保留不确定性，检验关键连接，明确责任，并依据经过验证的变化评价结果，而不是只看是否填满所有方框。