|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/tabtrainer-minitab-quality-charts-p-np-p-laneychart/
课程评论：没有评论
课程名称：Minitab中的P图与Laney P'图分析 - Tabtrainer 课程概述： 欢迎参加Tabtrainer®系列的高级课程，这是您在工业应用统计培训方面的可信来源。在本单元中，您将掌握在Minitab中使用P图和Laney P'图的技巧，重点是基于属性的过程控制，使用来自Smartboard公司的真实生产数据。课程内容超越了工具的使用，深入探讨了基础统计理论，包括二项分布、过散度/欠散度的诊断以及基于AIAG标准的控制限调整技术。此课程非常适合质量工程师、六西格玛专业人士和工业分析师，他们希望检测和解释过程不稳定性，了解子组大小变化的影响，并在实际生产环境中应用正确的控制图模型。 该培训由认证的TÜV培训师、Tabtrainer®创始人Murat Mola教授主导，确保课程既有技术深度又具工业相关性。Tabtrainer®专注于质量学习，提供经过验证的统计工具和清晰可行的见解。参与者不仅学习每种控制图的技术应用，还学习基础统计分布（例如二项分布和泊松分布）、诊断方法、过程不稳定性的解释，以及子组结构和过程变化对控制图准确性的影响。 完成此单元后，参与者将能够： - 理解Smartboard公司最终装配过程的结构和工作流程，包括早班、晚班和夜班。 - 依据表面检查结果分析滑板组件的分类系统，将其分为“良品”和“次品”属性类别，并识别因缺陷部件引起的返工和废品带来的财务后果。 - 导入和探索现实中的制造数据，数据包含装配日期、子组大小及整整一年的不良品数量（365个数据条目）。 - 认识名义尺度数据的特性，并理解其统计处理方法，特别是将二项分布应用于缺陷分类（“良品”vs“次品”）。 - 通过将不良滑板数量与每天的总生产量关联，计算每日的缺陷率。 - 进行全面的P图诊断，以验证现实中属性数据与二项分布理论期望的一致性，包括理解过散度和欠散度等概念。 - 解读概率图和一致性比率，以根据证据做出关于P图或Laney P'图适用性的决策。 - 创建和解读P图，展示不良滑板随时间变化的相对比例，基于不同的子组大小。 - 了解NP图的结构及其用途，NP图显示不良单位的绝对数量，并将其与P图进行比较。 - 正确识别和响应控制图测试所揭示的过程不稳定性（例如，通过测试1检测到的特殊原因 - 超出三个标准差的点）。 - 对发现的过程不稳定性进行根本原因分析，如假期相关的人员问题导致的缺陷率增加的实际示例。 - 根据AIAG指导原则评估在子组大小变化时平滑控制限的可能性，包括子组大小阈值计算和改进可解释性的调整。 - 理解子组大小变化对控制限和置信区间的影响，以及较大或较小的子组大小如何影响统计精度。 - 学习在AIAG标准允许的情况下，通过子组大小平均平滑控制限的实际步骤。 - 区分经典P图足够的情况与由于显著系统性散布效应需要使用Laney P'图的情况。 - 当P图诊断指示显著偏离二项散布假设时，创建和解读Laney P'图。 - 理解使用名义尺度属性数据进行质量控制和持续过程改进的实际意义。 - 根据现实中属性数据分析和控制图解释，判断制造过程是否可以认为是稳定的。 - 将完整的质量控制分析文档化，并以结构化项目格式（“缺陷率最终装配”）保存，以供后续参考和报告用途。
Welcome to this expert-level course from the Tabtrainer® Series - your trusted source for applied statistical training in industry.In this training unit, you will master the use of P Charts and Laney P' Charts in Minitab, with a focus on attribute-based process control using real-world production data from the Smartboard Company. Going beyond tool usage, this course dives into the underlying statistical theory, including binomial distribution, diagnostics for overdispersion/underdispersion, and advanced control limit adjustment techniques based on AIAG standards.This course is ideal for quality engineers, Six Sigma professionals, and industrial analysts who want to detect and interpret process instabilities, understand the effect of subgroup size variation, and apply the correct control chart models in real production settings.Developed and led by Prof. Dr. Murat Mola, certified TÜV trainer and founder of Tabtrainer®, this training guarantees both technical depth and industrial relevance. Tabtrainer® stands for quality-focused learning with proven statistical tools and clear, actionable insights.Participants learn not only the technical application of each control chart but also the underlying statistical distributions (binomial and Poisson), diagnostics, interpretation of process instabilities, and the impact of subgroup structure and process changes on control chart accuracy.After completing this training unit, participants will be able to:Understand the structure and workflow of the final assembly process at Smartboard Company, including early, late, and night shifts.Analyze the classification system for skateboard components based on surface inspection results into "good" and "bad" attribute categories, and recognize the financial consequences of rework and scrap due to defective parts.Import and explore real-world manufacturing data, consisting of assembly dates, subgroup sizes, and number of bad parts across a full year (365 data entries).Recognize the nature of nominally scaled data and understand its statistical treatment, particularly the application of binomial distribution for defect classification ("good" vs. "bad").Calculate daily defect rates by relating the number of bad skateboards to the total production volume per day.Perform a comprehensive P Chart Diagnostic to verify the conformity of real-world attribute data with the theoretical expectations of binomial distribution, including understanding concepts such as overdispersion and underdispersion.Interpret probability plots and agreement rates to make evidence-based decisions on the appropriate use of P Charts or Laney P' Charts.Create and interpret P Charts that visualize the relative proportion of defective skateboards over time, based on variable subgroup sizes.Understand the structure and use of NP Charts, which show the absolute number of defective units, and compare them to P Charts.Identify and correctly respond to process instabilities revealed through control chart tests (e.g., special causes detected via Test 1 - points outside three standard deviations).Perform root cause analysis for detected process instabilities, as demonstrated in the practical example of increased defect rates due to holiday-related staffing issues.Apply AIAG guidelines to assess the possibility of smoothing control limits when subgroup sizes vary, including calculations of subgroup size thresholds and adjustments for improved interpretability.Understand the impact of subgroup size variation on control limits and confidence intervals, and how larger or smaller subgroup sizes influence statistical precision.Learn the practical steps to apply smoothing of control limits by averaging subgroup sizes, when permitted under AIAG standards.Differentiate between situations where classical P Charts are sufficient and where Laney P' Charts are necessary due to significant systematic scatter effects.Create and interpret Laney P' Charts when P Chart diagnostics indicate significant deviations from binomial dispersion assumptions.Understand the practical implications of working with nominally scaled attribute data for quality control and continuous process improvement.Conclude whether a manufacturing process can be considered stable based on real-world attribute data analysis and control chart interpretation.Document and save the complete quality control analysis in a structured project format ("Defect Rate Final Assembly") for further reference and reporting purposes.
