Tutorial 1

3:30 to 5:00 p.m. Wednesday, October 13, 2010

Introduction to Design of Experiments

Design of Experiments (DOE) is a great tool to optimize products and processes for improving performance and reliability. The conventional one factor at a time (OFAT) strategy is inefficient and may not help you to get the best out of the processes. Statistically designed experiments are more efficient in terms of cost and effort and also enable us to reach the optimum setting faster. The tool is about 80 years old but is still not so widely used due to lack of knowledge and training. Modern software makes it extremely easy to plan and analyze experiments.

This tutorial discusses experimental strategy and some case studies in DOE and response optimization.

Key Words: Statistically Designed Experiments, Interactions, Orthogonal Arrays, Full Factorial Designs, Fractional Factorial Designs, Response Surface Designs, Robust Design

Hemant Urdhwareshe

Institute of Quality and Reliability