Skip to content

Latest commit

 

History

History
39 lines (19 loc) · 1.66 KB

README.md

File metadata and controls

39 lines (19 loc) · 1.66 KB

Experimental Design for Optimization in Manufacturing

Minimizing cost and data request runs for expensive trials by utilizing full factorial experiment to identify significance, centerpoints for investigating significance of curvature, and central composite design to measure curvature. In depth investigation of effects, significance, transformations, Analysis of variance, etc. Summary shown here. Details are in the report.

Step 1: Run a 2k factorial design

Run 2k factorial design to get following points:

alt text

Isolate the significant features:

alt text

Track results of regression model:

alt text

Step 2: Run center points design

Run centerpoints design on signficiant variables to get following points:

alt text

Track results of regression model:

alt text

Step 3: Run a central composite design

Run centerpoints design on signficiant variables to get following points:

alt text

Track results of regression model:

alt text