科研成果详情

Fractional factorial designs (FFDs) have received a significant attention in recent years due to their cost-effective and practical applicability to such diverse fields as medicine, agriculture, industry, and high-tech. While the research on regular FFDs arising from defining relations among active factors is now quite rich, recently, it has been increasingly recognized that nonregular FFDs can potentially perform even better. One of the significant obstacles for the use of nonregular FFDs is the lack of simple design structure. This paper explores the joint force potential of the multiple doubling technique (Elsawah, Stat Pap 62(6):2923–2967, 2021) and quaternary linear codes towards the construction of four-level nonregular FFDs that accommodate a large number of factors and hence are attractive from the practical viewpoint of experimental economy. A general simple systematic construction approach is given, and some theoretic results are obtained for the generated four-level nonregular FFDs to investigate their statistical properties in terms of the Hamming distance, Lee distance, alias structure, and power moment. Compared to the existing widely used techniques, the numerical results show that the new approach offers several advantages and its new generated FFDs perform better.