Statistical Tolerance

Introduction

This seminar will include a review of statistical theory and present statistical methods, which are used to better select and/or analyze Tolerance Stack-ups. The Probability (RMS) Method and tolerance optimization techniques will be discussed along with guidelines on which method(s) to use in given situations. Attendees will also view a demonstration of a excel based simulation program that analyzes the effects of form and assembly variation on the quality of a finished product.

Note: Participants should bring a scientific calculator for several in-class exercises

Objective

By attending in this seminar, you will be able to:

• Apply worst case, root-mean-square, and six sigma methods for the allocation of analysis of simple-to intermediate complexity tolerancing schemes
• Use the "Risk of Misassembly" approach for tolerance allocation, and the "Main Effect" approach for
• determining dimensional variables    tolerance which exhibit the greatest impact on build variation
• Understand and be exposed to the computer tools which can greatly improve their statistical tolerancing efforts, given the intricacies of GD&T, plus-minus tolerancing and various datum schemes

Course Content

• Review of Tolerancing Methods, Tolerance Stack-Ups and the relationship between Tolerancing and Quality. A High-Level Overview of Geometric Dimensioning & Tolerancing (GD&T) and Process Capability Measurement is Provided
• Tolerance Synthesis (Allocation) Versus Tolerance Analysis
• Overview of the Worst Case (non-statistical) Tolerancing Method for comparison with Statistical Tolerancing Results
• Probability & Statistics Concepts Required for Statistical Tolerancing Methods

1. Tolerance Allocation Based on "Risk of Misassembly"
2. Statistical Tolerancing Using the Root-Mean-Square (RMS) Method
3. With bilateral tolerances
4. With unilateral and/or asymmetrical tolerances
5. In 2-D and 3-D applications
6. Participant exercises

• Analyzing Part Tolerances using Main Effect and Sensitivity Analysis Methods - Methods for Determining the Contribution of Process Variables to Overall Process Variation
• Tolerance Optimization Techniques - Their Benefits in Effective Tolerancing of Parts and Assemblies
• Introduction to Monte Carlo Analysis

Target Audience

This seminar is intended for engineers  who would like to have a good working knowledge of applying statistics to product design in order to better predict and improve product quality