DOE Information

DOE - COURSE INFROMATION

(Web-Based : LAN-Based : CD-ROM)

Course Description
DOE: Screening Experiments are the most powerful of Design of Experiments techniques to uncover the power factors in a manufacturing process. With a small number of experimental runs and very little data your can make significant improvements to processes. The most common screening experiment techniques - Taguchi and Plackett-Burman techniques are covered in detail in this course.

This course includes three units: Background for DOE, Plackett-Burman Techniques, and Taguchi Techniques.

Intended Audience
Operators, engineers, supervisors, and managers who will be using control charts in their jobs.

Time To Complete
Approximately five to seven hours.

| overview | outline | why does my company need this program? |

COURSE OBJECTIVES

Unit 1: Background for DOE:

	Why design of experiments (DOE) are more efficient and effective than one-at-a-time (OAT) experimentation;
	What the major terms used in designed experiments mean;
	Types of designed experiments and when they are best used;
	How to use basic tests of significance;
	How to plan a designed experiment.

Unit 2-Plackett-Burman Techniques:

	How Plackett-Burman designs were derived;
	How to determine which Plackett-Burman matrix to use for your application;
	How to estimate the experimental error in Plackett-Burman experiments;
	How to analyze experimental results and calculate the statistical significance of factor effects;
	How to develop a prediction equation that can be used to optimize the response;
	How to salvage experiments if data are lost.

Unit 3-Taguchi Techniques:

	Why Taguchi techniques focus on the robustness of the product;
	How the Quality (Taguchi) Loss Function is used;
	How to calculate and use Signal to Noise ratios;
	How Taguchi designs were derived;
	How to determine which Taguchi design to use for your application;
	How to use Taguchi interaction tables;
	How to test the statistical significance of factor effects;
	How to develop a prediction equation that can be used to optimize the response;
	How to use mean, S/N, and variation effects to determine where to set factors;
	How to salvage experiments if data are lost.