质量的七个基本工具

Metropolis-Hastings算法是一种著名的马尔可夫链蒙特卡罗（MCMC）方法，用于从难以直接采样的概率分布中获取随机样本序列。在每次迭代中，它基于当前样本生成下一个样本的候选值。然后，该候选值以一定的概率被接受或拒绝，从而确保生成的链收敛到期望的分布。

面向对象编程 (OOP) 中的抽象是指隐藏复杂的实现细节，仅显示对象的基本功能。它关注的是对象做什么，而不是如何做。抽象类和接口可以实现这一点，它们为其他类定义蓝图，但不提供完整的实现，从而简化了复杂的系统。

瀑布模型是一种顺序的、非迭代的软件开发过程，其进度像瀑布一样，通过不同的阶段稳步向下流动：构思、启动、分析、设计、构建、测试、部署和维护。每个阶段都必须完全完成才能进入下一个阶段。瀑布模型通常与迭代模型形成对比，以突出其灵活性。

恢复块方案是一种基于设计多样性和向后错误恢复的软件容错技术。它将程序结构化为一系列块，每个块包含一个主模块、一个验收测试以及一个或多个备用模块。如果主模块的输出未能通过验收测试，则系统状态恢复，并执行一个备用模块。

Verification and validation (V&V) are distinct processes. Verification ensures a product meets its specified requirements ("Are you building it right?"). Validation ensures the product meets the user's actual needs and intended use ("Are you building the right thing?"). They are complementary activities within quality management, often performed sequentially or in parallel to ensure both correctness and usefulness.

Verification techniques are broadly classified as static or dynamic. Static verification (or static analysis) examines the system's code or design without executing it. Examples include code reviews, inspections, and automated static analysis tools. Dynamic verification (or testing) involves executing the system with a set of inputs and observing its behavior to find defects. Both are complementary for comprehensive quality assurance.

The Risk Priority Number (RPN) is a quantitative measure used in FMEA to prioritize risks. It is calculated as the product of three ranked factors: Severity (S), Occurrence (O), and Detection (D). The formula is [latex]RPN = S times O times D[/latex]. Each factor is typically rated on a scale from 1 to 10, allowing teams to focus on the highest-scoring risks first.

根据错过截止时间的后果，实时系统可分为“硬实时”和“软实时”。在硬实时系统中，错过截止时间会导致系统完全失效，例如防抱死制动系统。在软实时系统中，错过截止时间会导致性能下降，但不会导致灾难性故障，例如实时音视频流。

速率单调调度（RMS）是一种用于实时系统中周期性任务的静态优先级调度算法。它根据任务频率分配优先级：任务周期越短（速率越高），优先级越高。RMS 是一种最优静态优先级算法，这意味着如果任何静态优先级算法能够调度一组任务，RMS 也能够调度。可以通过基于利用率的测试来检验其可调度性。

有限体积法 (FVM) 是 CFD 中用于求解偏微分方程的主要数值方法。它将计算域离散为控制体网格，并将控制方程的积分形式应用于每个控制体。通过使用散度定理将体积积分转换为表面积分，FVM 专注于计算跨单元表面的守恒性质的通量。

Formal verification is the use of mathematical methods to prove or disprove the correctness of a system's design with respect to a formal specification. Unlike testing, which can only show the presence of bugs for specific inputs, formal verification can prove their absence for all possible inputs. It involves creating a formal model of the system and using techniques like model checking or theorem proving.