Workers’ Control: How Control Charts Empower Workers in Quality Assurance | 工人控制：控制图如何赋能质量保证中的工人

📚 Workers’ Control: How Control Charts Empower Workers in Quality Assurance | 工人控制：控制图如何赋能质量保证中的工人

In modern manufacturing, workers are not just machine operators; they are the front-line guardians of product quality. Through statistical process control, employees use control charts to monitor production processes, detect issues early, and maintain high standards. This article explores how workers’ control, rooted in Edexcel A‑level Mathematics, transforms data into actionable insights.

在现代制造业中，工人不仅是机器操作者，更是产品质量的一线守护者。通过统计过程控制，员工利用控制图监控生产过程，及早发现问题，维持高标准。本文基于Edexcel A‑level 数学，探讨工人控制如何将数据转化为可执行的洞察。

---

1. Introduction to Workers’ Control and Quality Assurance | 工人控制与质量保证简介

Workers’ control refers to the active participation of shop‑floor operators in monitoring and improving quality using statistical tools. Instead of relying solely on dedicated inspectors, companies empower workers to record measurements, plot control charts, and signal when the process drifts out of control. This approach not only boosts efficiency but also gives employees a deeper sense of ownership.

工人控制指的是车间操作员利用统计工具积极参与监控和改进质量。公司不再仅依赖专职检验员，而是授权工人记录测量值、绘制控制图，并在过程失控时发出信号。这种做法不仅提高效率，还赋予员工更强的归属感。

At the heart of this empowerment lies the control chart, a foundational concept in Edexcel A‑level Statistics modules. The chart translates sampled data into a visual tool that distinguishes natural variation from genuine process shifts.

这种赋权的核心正是控制图，这是 Edexcel A‑level 统计学模块中的基础概念。控制图将样本数据转化为一种可视化工具，能区分自然变异与真正的过程偏移。

---

2. The Role of Control Charts in Manufacturing | 控制图在生产中的角色

A control chart is a time‑ordered plot of sample statistics, such as the sample mean or sample range, drawn with a central line and two pairs of limits: warning limits and action limits. Workers use these charts to judge whether a process is stable and predictable. If points fall outside the limits or show unusual patterns, the worker can stop the line before defective units pile up.

控制图是按时间顺序绘制的样本统计量图，例如样本均值或样本极差，图中配有中心线以及两对界限：警告界限和行动界限。工人利用这些图表判断过程是否稳定、可预测。如果数据点落在界限之外或呈现异常模式，工人可在不合格品大量堆积之前停机。

In the Edexcel syllabus, control charts are taught as an application of the normal distribution and hypothesis testing. This directly connects the mathematical theory with real‑world decision‑making on the factory floor.

在 Edexcel 大纲中，控制图作为正态分布和假设检验的应用来教授。这直接将数学理论与工厂车间的实际决策联系起来。

---

3. Understanding Variation: Common and Special Causes | 理解变异：普通原因与特殊原因

Every manufacturing process exhibits variation. Common cause variation is inherent, random, and arises from the system itself – slight temperature changes, raw material inconsistencies, or machine vibrations. Special cause variation, on the other hand, is assignable: a tool wearing out, an incorrect setting, or a new batch of faulty ingredients.

每个生产过程都会呈现变异。普通原因变异是固有的、随机的，来自系统本身——轻微的温度变化、原材料的不一致或机器振动。另一方面，特殊原因变异是可归因的：刀具磨损、设置错误或一批新的劣质原料。

Workers’ control relies on the ability to distinguish these two types. A control chart sets statistical boundaries so that workers can tell when a process is merely ‘noisy’ and when it truly requires intervention. This prevents unnecessary adjustments that can actually increase variability.

工人控制依赖于区分这两种类型的能力。控制图设定了统计边界，使工人能够分辨过程仅仅是“有噪声”还是真正需要干预。这避免了不必要的调整，因为不必要的调整反而会增加变异性。

---

4. The X‑bar Chart: Monitoring the Process Mean | 均值图：监控过程均值

The X‑bar chart (mean chart) tracks the sample means of small subgroups taken at regular intervals, for example, five items every half hour. Under the assumption that the process mean µ and standard deviation σ are known, the sample mean X̄ follows a normal distribution with mean µ and standard error σ/√n.

均值图追踪每隔固定时间抽取的小子组的样本均值，例如每半小时抽取五件产品。在已知过程均值 µ 和标准差 σ 的假设下，样本均值 X̄ 服从正态分布，其均值为 µ，标准误为 σ/√n。

Workers plot each X̄ value in time order. The central line is set at the target mean µ. If the process is in control, about 99.7 % of sample means should lie within three standard errors of µ, providing a clear visual benchmark.

工人按时间顺序标绘每个 X̄ 值。中心线设在目标均值 µ 处。如果过程受控，大约 99.7 % 的样本均值应落在 µ 的三个标准误范围内，这提供了清晰的视觉基准。

---

5. Setting Control Limits: Action and Warning Limits | 设定控制界限：行动界限与警告界限

In Edexcel A‑level mathematics, control limits are derived from the normal distribution. The warning limits are typically set at µ ± 2σ/√n, because only about 5 % of values fall outside this interval when the process is on target. The action limits are set at µ ± 3σ/√n, corresponding to a single‑tail probability of roughly 0.1 % above or below. A point beyond an action limit strongly suggests the process mean has shifted.

在 Edexcel A‑level 数学中，控制界限由正态分布导出。警告界限通常设为 µ ± 2σ/√n，因为过程在目标状态时，只有约 5 % 的数值会落在此区间之外。行动界限设为 µ ± 3σ/√n，对应的单侧概率大约为 0.1 %。若一点超出行动界限，则强烈表明过程均值已发生偏移。

Limit type	Formula	Significance
Upper Action Limit (UAL)	µ + 3σ/√n	1 observation beyond → investigate immediately
Upper Warning Limit (UWL)	µ + 2σ/√n	2 of 3 consecutive points beyond → possible shift
Lower Warning Limit (LWL)	µ − 2σ/√n	Symmetric to upper warning
Lower Action Limit (LAL)	µ − 3σ/√n	Immediate action if breached

界限类型 | 公式 | 意义
上行动界限 UAL | µ + 3σ/√n | 一点超出 → 立即调查
上警告界限 UWL | µ + 2σ/√n | 连续三点中有两点超出 → 可能偏移
…（表格对应翻译）

These limits transform statistical theory into clear rules. Workers need not compute probabilities every time; they simply check the chart against known thresholds.

这些界限将统计理论转化为清晰的规则。工人无需每次都计算概率；只需对照已知阈值检查图表即可。

---

6. The R Chart: Monitoring Process Variability | 极差图：监控过程变异

While the X‑bar chart checks the process centre, the R chart (range chart) tracks the spread or variability within each subgroup. The sample range R is the difference between the largest and smallest observation in a subgroup. A stable process must show consistent variability; an increasing range can indicate deteriorating equipment or inconsistent materials.

均值图检查过程的中心，而极差图则追踪每个子组内的分散程度或变异性。样本极差 R 是子组中最大值与最小值之差。稳定的过程必须呈现一致的变异性；增大的极差可能表示设备劣化或材料不一致。

To construct an R chart, workers calculate the mean range R̄ from past data and, using statistical tables or constants, set upper and lower control limits. In Edexcel contexts, the standard error of the range is estimated, and limits are often expressed as D₄R̄ (upper) and D₃R̄ (lower), where D₃ and D₄ depend on subgroup size n. Points outside these limits signal that the process variation is no longer at its usual level.

为构建极差图，工人根据历史数据计算平均极差 R̄，并利用统计表或常数设定上下控制界限。在 Edexcel 范围内，极差的标准误通过估计得出，界限通常表示为 D₄R̄（上界限）和 D₃R̄（下界限），其中 D₃ 和 D₄ 取决于子组大小 n。若有点落在界限之外，则表明过程变异已偏离正常水平。

---

7. How Workers Interpret Control Charts | 工人如何解读控制图

Effective workers’ control depends on simple, teachable decision rules. The chart itself does not make decisions; the worker does. Standard rules include: one point outside an action limit requires immediate investigation; two out of three successive points beyond a warning limit on the same side of the centre suggest a sustained shift; and a run of seven points all above or all below the centre line can also indicate a process mean change.

有效的工人控制依赖于简单、可传授的决策规则。图表本身不做决策，工人做决策。标准规则包括：一点超出行动界限则需立即调查；连续三点中有两点在中心线的同一侧超出警告界限提示发生持续偏移；连续七点全部在中心线之上或之下也可能表明过程均值发生了变化。

These rules are direct applications of the binomial and normal distributions taught in A‑level Statistics. For example, the probability of seven consecutive points lying on the same side of the centre by chance is (½)⁷ = 1/128 ≈ 0.0078, which is highly unlikely, so the worker is justified in suspecting a genuine shift.

这些规则是 A‑level 统计学中教授的二项分布与正态分布的直接应用。例如，连续七点偶然落在中心线同一侧的概率为 (½)⁷ = 1/128 ≈ 0.0078，这极不可能发生，因此工人有充分理由怀疑存在真实的偏移。

---

8. Hypothesis Testing in Control Charts | 控制图中的假设检验

Every time a worker compares a newly plotted point against a control limit, they are informally performing a hypothesis test. The null hypothesis H₀ states that the process mean is equal to the target µ. The alternative hypothesis H₁ is that the mean has shifted. An out‑of‑control signal corresponds to rejecting H₀ at a certain significance level.

每当工人将新绘制的点与控界限进行比较时，他们都在非正式地进行一次假设检验。原假设 H₀ 为过程均值等于目标 µ，备择假设 H₁ 为均值发生了偏移。失控信号相当于在一定显著性水平下拒绝 H₀。

If the warning limits use ±2 standard errors, the significance level for one point outside is approximately 5 % (two‑tailed). With action limits at ±3 standard errors, the significance level drops to about 0.27 %, meaning that if the process is truly on target, only one in about 370 samples would falsely trigger an action alarm. This trade‑off between false alarms and missed shifts is at the core of statistical quality control.

若警告界限使用 ±2 个标准误，单点位于界限之外的显著性水平约为 5 %（双尾）。行动界限设在 ±3 个标准误时，显著性水平降至约 0.27 %，这意味着如果过程确实在目标状态，大约每 370 个样本中才会有一个错误地触发行动警报。这种误报与漏报之间的权衡正是统计质量控制的核心。

---

9. Practical Example: Workers Using Control Charts on a Production Line | 实际案例：工人在生产线上使用控制图

A biscuit factory packs cookies into 200 g packets. The known process standard deviation σ is 2.5 g, and the target mean µ = 200 g. Workers draw samples of n = 5 packets every 20 minutes and calculate the sample mean. The standard error is σ/√5 ≈ 1.118 g. Therefore, the action limits are 200 ± 3×1.118, i.e. 196.65 g and 203.35 g; the warning limits are 200 ± 2×1.118, i.e. 197.76 g and 202.24 g.

一家饼干厂将饼干装入 200 g 的包装袋。已知过程标准差 σ 为 2.5 g，目标均值 µ = 200 g。工人每 20 分钟抽取 n = 5 的一个样本并计算样本均值。标准误为 σ/√5 ≈ 1.118 g。因此，行动界限为 200 ± 3×1.118，即 196.65 g 和 203.35 g；警告界限为 200 ± 2×1.118，即 197.76 g 和 202.24 g。

At 10:00 am, a sample gives a mean of 203.8 g – above the upper action limit. The operator immediately halts the filling machine, finds that a measuring cup has loosened, and fixes it. Without this workers’ control intervention, thousands of overfilled packets could have reached customers, wasting product and breaching regulations.

上午 10:00，一个样本的均值为 203.8 g——超出上行动界限。操作员立即停止灌装机，发现一个量杯松动了，于是将其修好。若没有这种工人控制干预，成千上万包超重的产品就可能送到顾客手中，既浪费产品又违反规定。

---

10. Estimating σ from the Sample Range | 从样本极差估计 σ

In many real processes, the true process standard deviation σ is unknown and must be estimated from pilot data. A common method uses the average range R̄. For a stable subgroup size n, the standard deviation can be approximated by σ̂ = R̄ / d₂, where d₂ is a constant found in statistical tables. For n = 5, d₂ ≈ 2.326.

在实际过程中，真实的过程标准差 σ 通常是未知的，必须根据初始数据来估计。一种常见的方法是使用平均极差 R̄。对于固定的子组大小 n，标准差可通过 σ̂ = R̄ / d₂ 来近似，其中 d₂ 是统计表中的常数。当 n = 5 时，d₂ ≈ 2.326。

Workers then compute control limits for the X‑bar chart using this estimate: UAL/LAL = µ ± 3(R̄/d₂)/√n. This links the R chart and the X‑bar chart, allowing workers to update limits as the process improves, without needing in‑depth statistical calculations on the shop floor.

然后，工人利用该估计值计算均值图的控制界限：UAL/LAL = µ ± 3(R̄/d₂)/√n。这便将极差图与均值图联系起来，使得工人能够在过程改进时更新界限，而无需在车间进行复杂的统计计算。

---

11. Benefits and Limitations of Workers’ Control with Control Charts | 工人使用控制图的优势与局限

Benefits include early detection of problems, reduced scrap and rework, improved communication between shifts, and a culture of shared responsibility. Workers become active problem‑solvers rather than passive executors of instructions. However, limitations exist: poor training can lead to misinterpretation, and some special causes require engineering expertise beyond the scope of the shop‑floor worker. Additionally, if processes change frequently, static control limits may become outdated.

优势包括尽早发现问题、减少废品和返工、改善班组间的沟通以及培养共同责任的文化。工人成为主动解决问题者，而非被动的指令执行者。然而，也存在局限性：培训不到位可能导致误判，一些特殊原因需要的工程专业知识超出车间工人的能力范围。此外，如果过程频繁变化，静态控制界限可能会过时。

The Edexcel specification highlights the importance of regularly updating control limits and training workers to understand both the mathematical basis and the practical implications of the charts they use.

Edexcel 大纲突出了定期更新控制界限以及培训工人的重要性，使工人既理解所用图表的数学基础，也明白其实际意义。

---

12. Conclusion | 结论

Workers’ control, powered by statistical control charts, is a practical embodiment of A‑level Mathematics in the working world. By equipping operators with the ability to monitor and respond to variation, manufacturers achieve higher quality at lower cost. The concepts of means, standard errors, probability, and hypothesis testing are not just exam topics; they are daily tools in the hands of informed workers.

以统计控制图为支撑的工人控制，是 A‑level 数学在职场中的具体体现。通过赋予操作员监控和应对变异的能力，制造商以更低的成本实现了更高的质量。均值、标准误、概率和假设检验等概念不仅是考试内容，也是掌握知识的工人手中的日常工具。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）