故事 · 从摩托罗拉车间到韦尔奇的 120 亿
Origin Story · From a Motorola Floor to Welch's $12 Billion
1986 年,摩托罗拉的工程师 Bill Smith 提出一个朴素却革命性的想法:与其等产品做坏了再返工,不如把过程本身做到几乎不出错。
他给这个目标起名「六西格玛」。真正让它名扬天下的,是 1995 年 GE 的 CEO 杰克·韦尔奇 —— 他不把它当成质量部的工具,而是当成全公司的战略,
从财务到客服全员推行,并用 DMAIC(定义-测量-分析-改进-控制)把每个改善都做成可复制的项目。几年间,GE 报告省下约 120 亿美元。
它的核心洞见只有一句:缺陷不是质量部的小问题,是吃掉公司利润的大黑洞 —— 把质量从「成本中心」翻转成「利润引擎」。
In 1986 Motorola engineer Bill Smith made a humble but revolutionary suggestion: instead of fixing bad parts after the fact, build a process that almost never produces them.
He named the goal Six Sigma. What made it world-famous was 1995, when GE's CEO Jack Welch treated it not as a QC tool but as a company-wide strategy —
rolling it from finance to customer service, with DMAIC (Define-Measure-Analyze-Improve-Control) turning every improvement into a repeatable, auditable project. In a few years GE reported roughly $12 billion in savings.
The core insight is one line: defects aren't a QC problem — they're a black hole eating company profit. Six Sigma flipped quality from "cost center" to "profit engine".
1 看不见的隐藏工厂:缺陷在偷你的利润 The Invisible Hidden Factory: Defects Stealing Profit
隐藏工厂巨大Hidden factory is huge2 σ 每升一级,质量成本断崖式下跌 Each σ Level Cuts Quality Cost off a Cliff
橙线是劣质成本占营收比(COPQ%):3σ 企业约烧掉营收的 25~40%,而 6σ 企业不到 1%。每提升一个 σ,缺陷几乎缩小一个数量级 —— 这就是「四个 9 冲到六个 9」如此值钱的原因。 The orange line is Cost of Poor Quality as % of revenue. A 3σ company burns 25–40%; a 6σ company burns under 1%. Each σ level shrinks defects by roughly an order of magnitude — that's why "moving from four-nines to six-nines" is worth so much.
3 现实里的六西格玛 Six Sigma in the Real World
管理哲学:它不是质量部的私房工具,而是一套全员、用数据说话、以客户为中心的经营方式。韦尔奇把它写进每个经理的考核。
A management philosophy: not a QC team's secret tool, but a company-wide, data-driven, customer-centric way of running the business. Welch tied it to every manager's bonus.
DMAIC 流程:定义-测量-分析-改进-控制。它把「拍脑袋救火」变成可复制、可验证、不复发的项目化改善。
The DMAIC loop: Define-Measure-Analyze-Improve-Control. It turns gut-feel firefighting into projectized improvement that's repeatable, verifiable, and stays fixed.
隐藏工厂:返工、检验、报废、投诉、退货 —— 这些劣质成本(COPQ)常占营收 15~40%,却从不出现在任何一张产能报表上。
The hidden factory: rework, inspection, scrap, complaints, returns. This Cost of Poor Quality routinely runs 15–40% of revenue — yet shows up on zero capacity reports.
GE 与摩托罗拉:摩托罗拉 1988 年首夺鲍德里奇国家质量奖;GE 几年省下约 120 亿美元 —— 真金白银证明质量是利润。
GE & Motorola: Motorola won the inaugural Baldrige National Quality Award in 1988; GE banked roughly $12 billion in a few years — hard proof that quality is profit.
一句话In One Line
六西格玛真正颠覆的,不是统计,而是看待质量的视角。传统观念把质量看成「多花的钱」—— 多检验、多返工、多保固;
六西格玛说反了:每一个缺陷都在悄悄抽走你已经赚到的利润。把过程做到 6σ,等于关掉那座没人统计的隐藏工厂,省下的钱直接落进利润表。
而 DMAIC 让这件事不靠英雄式救火,而靠人人可学、可复制、可验证的方法。这就是为什么它能从一家芯片厂的车间,长成横扫全球的管理革命。
What Six Sigma really overturned isn't statistics — it's the way you see quality. The old view called quality "extra money spent" — more inspection, more rework, more warranty.
Six Sigma flipped it: every defect is quietly siphoning profit you've already earned. Running a process at 6σ shuts down that uncounted hidden factory, and the savings drop straight to the bottom line.
DMAIC makes the work survive without heroics — teachable, repeatable, verifiable. That's how a method from a chip plant's shop floor grew into a global management revolution.
常见误用Common Mistakes
把六西格玛当成纯统计培训,拿了证书却不动业务。它首先是经营战略,要由管理层挂帅、对准真实痛点立项。
Treating Six Sigma as a stats training program — collect the belt, change nothing in the business. It's a business strategy first. Executive sponsorship and projects aimed at real pain are non-negotiable.
只盯产线良率,忽视服务与流程。订单、报价、客服等非制造流程同样有隐藏工厂,省钱空间常更大。
Only looking at line yield, ignoring services and back-office processes. Order entry, quoting, customer service all have their own hidden factories — often with even bigger savings.
为追 6σ 而过度投入、得不偿失。目标 σ 水平要按缺陷的经济后果定,安全件求高、非关键件够用即可。
Chasing 6σ everywhere at any cost. Target σ level should match the economic consequence of failure — safety-critical items go high, non-critical items go "good enough".