Interaction Graphs For A Two-Level Combined Array Experiment Design

oped the undergraduate and graduate programs in Statistics and has taught a variety of courses in Writing. Dr. Walker's industrial experience includes 12 years with airborne weapons systems integration and automation, supervision and administration, project management, and program management in countries throughout the Pacific Rim, Australia, and Africa. Fred's consulting experience includes ten years of continual involvement with internationally-based manufacturers from electronics, pulp and paper, bio-Abstract In planning a 2 k-p fractional factorial experiment, prior knowledge may enable an experimenter to pinpoint interactions which should be estimated free of the main effects and any other desired interactions. Taguchi (1987) gave a graph-aided method known as linear graphs associated with orthogonal arrays to facilitate the planning of such experiments. Since then the graph-aided method has been enhanced by various authors such as Li et al (1991), Wu and Chen (1992), Robinson (1993), and Wu and Hamada (2000). In this paper, we propose an algorithm for developing all possible non-isomorphic interaction graphs for combined arrays.