Constructing General Orthogonal Fractional Factorial Split-Plot Designs

• Activity Number: 329
• Type: Invited
• Date/Time: Tuesday, August 6, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Eric D Schoen*+ and Bagus Sartono and Peter Goos
• Companies: TNO, Dept. Quality and Safety and IPB and University of Antwerp
• Keywords: linear programming ; mixed-level design ; multi-level design ; orthogonal array ; two-level design ; variable neighborhood search

Abstract

There are a few major voids in the literature on orthogonal design of fractional factorial split-plot (FFSP) experiments. First, designs with one or more factors acting at more than two levels have not yet been considered. Second, published work on nonregular FFSP designs was either based only on Plackett-Burman designs, or on nonregular designs with limited numbers of factors. In this talk, we present a novel approach to designing general orthogonal FFSP designs. It can handle two-level designs as well as designs involving one or more factors with more than two levels. Moreover, the approach can be used to create two-level designs that match or outperform alternative designs in the literature, and to create two-level designs that cannot be constructed using existing methodology. Our new approach involves the use of integer linear programming and mixed integer linear programming. For large design problems, it combines integer linear programming with variable neighborhood search. We demonstrate our approach by means of 48-run and 96-run two-level designs, an 81-run three-level design, and a 48-run mixed-level design.