Duration
2 days
Audience
Engineers, managers and other technical personnel who will be involved in designing experiments for process improvement.
Objectives
- Identify best materials, design, methods and conditions that will optimize product performance or yield.
- Reduce process variation by controlling the correct key process input variables (KPIVs).
- Understand how fluctuations in KPIVs affect process output or performance.
Course Content
In this step-by-step, hands-on DOE workshop, you will learn:
- Brief history and principles of DOE
- Why is DOE such a powerful process improvement tool.
- Prerequisites of performing DOE.
- DOE vs the traditional one-factor-at-a-time (OFAT) model – how DOE achieves more by predicting interactions and reducing experimental time.
- Effective planning for DOE
- Establish problem situation and set objectives / targets to achieve.
- Select the input variables, x1,x2, x3,…,xn and their factor levels.
- Select appropriate experimental design and run DOE.
- Conduct experimental runs, collect and analyse data.
- Calculate main effects and effects of interaction.
- Draw conclusions and decide on next course of action.
- Use of graphical illustration to assist in data analysis and interpretation.
- Select most appropriate experimental design
- Select the number of factors and number of levels of experimental design.
- Pros and cons of full factorial vs fractional factorial design.
- Assess the need of randomization, replication and repetition in a DOE.
- Application of centre points to test linearity of a relationship.
- Application of confounding to negate the influence of unwanted/uncontrollable conditions.
- Taking calculated risks by using Plackett-Burman Designs to screen large number of KPIVs.
- Additional tools to facilitate implementation of DOE.
- Multi-vari charts to illustrate interaction in input variables.
- Multiple regression analysis as an estimator of process output when input variable values are known.
- Residual analysis and normal probability plot of residuals to detect any unaccounted for situations.
- Effective data collection methods to collect unbiased data.
- Hands-on application of DOE
- Interpret case studies on DOE.
- Design your own experimental runs.
- Conduct a follow-up experimental design based on given data.
- Avoid the 10 most common mistakes in DOE.
Programme scope, delivery arrangements and suitability are confirmed during enquiry.