|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/tabtrainer-minitab-capability-analysis-non-normal-data/
课程评论：没有评论
课程名称：使用 Minitab 进行高级过程能力分析 概述：欢迎参加本课程，这是 Tabtrainer® 系列的高级培训，专注于工业和学术界高影响力的统计培训。本课程由 Tabtrainer® 创始人、TÜV 认证的穆拉特·摩拉教授（Prof. Dr. Murat Mola）教授，他在德国获得了“2023 年度杰出教授”奖。Tabtrainer® 的课程以清晰、精确和可操作的结果著称，旨在弥合理论与工业应用之间的差距。 课程描述： 该综合性的三部分培训课程使参与者具备在多种现实制造场景中进行过程能力分析的基本知识和应用技能。利用统计质量控制的强大工具 Minitab，参与者将学习如何评估非正态分布的连续数据、二项（属性）数据和泊松分布的缺陷数据的过程性能。 课程包括三个模块，结合 Smartboard 公司的案例研究，不仅涵盖了每种分析类型背后的统计理论，还特别侧重于实践中的 Minitab 应用，确保质量工程的可靠和可操作洞察。 课程结构与学习成果： **第一部分 - 非正态数据的过程能力** 场景：滑板轴的热处理过程中的尺寸变化 学习内容： - 了解在非正态分布下经典过程能力指标（Cp、Cpk、Pp、Ppk）的局限性。 - 使用 Minitab 的描述性统计、箱型图及安德森-达林检验评估分布假设。 - 应用 Minitab 的个体分布识别确定最佳拟合转换模型。 - 执行 Johnson 转换，并通过 p 值和概率图进行验证。 - 利用 Minitab 的能力六袋分析过程稳定性和能力。 - 解读 Pp 和 Ppk 值以确定符合客户规范的限度。 **第二部分 - 二项分布数据的能力分析** 场景：最终组装滑板的表面检查 学习内容： - 区分名义比例数据与度量数据，理解其对统计建模的影响。 - 使用 Minitab 的 p 图评估在变动的子组大小下的过程稳定性。 - 验证二项分布假设，并进行能力分析，解读缺陷率、PPM 和 Z 基准等关键指标。 **第三部分 - 泊松分布缺陷计数的能力分析** 场景：最终装配中每个子组的表面划痕数 学习内容： - 理解何时以及为何对缺陷计数数据应用泊松分布。 - 使用 Minitab 的 U 图建模每单位缺陷数（DPU）。 - 分析过程稳定性并验证泊松分布假设，解释统计摘要如均值 DPU 和置信区间。 软件重点：整个课程中，Minitab 是中心工具。参与者将掌握数据导入与结构、分布识别与转换、控制图的选择与解读、能力分析工具（六袋、二项、泊松）以及理解报告和图形输出以供决策。 本课程致力于为参与者提供深入的统计实用技能，以支持质量工程和过程改进的决策。
Advanced Process Capability Analysis Using Minitab: From Non-Normal Data to Attribute Metrics (Binomial & Poisson) Welcome to this advanced training from the Tabtrainer® Series - a recognized learning platform for high-impact statistical training in industry and academia.This course is developed and taught by Prof. Dr. Murat Mola, founder of Tabtrainer®, certified by TÜV and awarded "Professor of the Year 2023" in Germany. Tabtrainer® courses are known for bridging the gap between theory and industrial application - with clarity, precision, and actionable outcomes.Course DescriptionPart 1 - Process Capability for Continuous Non-Normal DataPart 2 - Capability Analysis for Binomially Distributed Data (Good/Bad Classification)Part 3 - Capability Analysis for Poisson-Distributed Defect CountsCourse Description:This comprehensive three-part training course equips participants with the essential knowledge and applied skills to perform process capability analysis across a broad spectrum of real-world manufacturing scenarios. Using Minitab, one of the most powerful tools for statistical quality control, participants will learn how to evaluate process performance for non-normally distributed continuous data, binomial (attribute-based) data, and Poisson-distributed defect data.Through three hands-on modules based on the Smartboard Company case studies, the course not only covers the statistical theory behind each analysis type but also focuses on practical Minitab applications that ensure reliable and actionable insights in quality engineering.Course Structure and Learning Outcomes:Part 1 - Process Capability for Continuous Non-Normal DataScenario: Dimensional changes during heat treatment of skateboard axlesKey Learnings:Understand the limitations of classical process capability metrics (Cp, Cpk, Pp, Ppk) when data are non-normally distributed.Use Minitab's Descriptive Statistics, Boxplots, and the Anderson-Darling test to evaluate distribution assumptions.Apply Minitab's Individual Distribution Identification to determine the best-fitting transformation model.Perform Johnson Transformation and validate it via p-values and probability plots.Utilize the Capability Sixpack (Normal) in Minitab to analyze process stability and capability after transformation.Interpret Pp and Ppk values to determine conformance to customer specification limits.Conclude on process centering potential and sigma level adequacy (Six Sigma benchmark).Heavy emphasis on real-world data preprocessing and transformation in Minitab before performing capability analysis.Part 2 - Capability Analysis for Binomially Distributed Data (Good/Bad Classification)Scenario: Surface inspection of final assembled skateboardsKey Learnings:Differentiate nominal scale data from metric data and understand its implications on statistical modeling.Use Minitab's p-charts to evaluate process stability with respect to fluctuating subgroup sizes.Understand how to verify binomial distribution assumptions using rate of defectives plots, histograms, and cumulative defect curves.Conduct Capability Analysis for Binomial Data in Minitab using:"Statistics > Quality Tools > Capability Analysis > Binomial"Specification of varying or constant subgroup sizesInterpret key indicators such as defect rate, PPM, and Z benchmark.Evaluate whether the current process meets the Six Sigma threshold (Z ≥ 2.0).Participants gain hands-on skills to analyze binary attribute data (pass/fail, good/bad) using Minitab's specialized capability tools.Part 3 - Capability Analysis for Poisson-Distributed Defect CountsScenario: Number of surface scratches per subgroup in final assemblyKey Learnings:Understand when and why to apply Poisson distribution for defect count data.Learn to model defects per unit (DPU) using U-charts in Minitab.Use Minitab's "Capability Analysis > Poisson" function to evaluate process performance for count data.Analyze process stability using U-charts and cumulative DPU plots.Verify Poisson distribution assumptions using Poisson plots.Interpret summary statistics such as:Mean DPUExpected vs. observed defect levelsConfidence intervalsDerive improvement actions when current processes are not capable (e.g., scratch rate > 0%).This module focuses on translating real-time count data into statistically valid process insights using Poisson capability models in Minitab.Software Focus:Throughout all three modules, Minitab is the central tool. Participants will become proficient in:Data importing and structuringDistribution identification and transformationSelection and interpretation of control chartsCapability analysis tools (Sixpack, Binomial, Poisson)Understanding reports and graphical outputs for executive decision-making
