• Communications for Statistical Applications and Methods
• /
• v.9 no.2
• /
• pp.347-354
• /
• 2002

An n-point design is maximal fan if all the models with n-terms satisfying the divisibility condition are estimable. Such designs tend to be space filling and look very similar to the ″Latin-hypercube″ designs used in computer experiments. Caboara, Pistone, Riccomago and Wynn (1997) conjectured that a maximal fan design on an integer grid exists for any n and m, where m is the number of factors. In this paper we examine the relationship between maximal fan design and latin-hypercube to give a partial solution for the conjecture.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.707-714
• /
• 2003

Because of complex aliasing, nonregular designs have traditionally been used for screening only main effects. However, complex aliasing actually may allow some interactions entertained and estimated without making additional runs. According to hierarchical principle, the minimum aberration has been used as an important criterion for selecting regular fractional factorial designs. The criterion is not applicable to nonregular designs. In this paper, we give a criterion for choosing fractional factorial designs based on the fan theory. The criterion is focused on the partial ordering given by set inclusion on estimable sets which is called leaves.

• Communications for Statistical Applications and Methods
• /
• v.6 no.3
• /
• pp.707-718
• /
• 1999

Non-regular designs such as the Plackett-Burman(PB) design have traditionally been used for screening only main effects because of complex aliasing. But it was found that these designs could be used to estimate the 2-factor interactions as well as main effects through the hidden projection property. The goal of this paper is to propose the estimable model when projecting the 12-run PB design using the algebraic geometric method. The core of this method considers the design as a affine variety and the Grbner basis of the design ideal for this affine variety gives the estimable polynomial models. As the results of applying the 12-run PB design it is actually found that this design has the models not only with 2-factor interactions but with 3-factor. This design is the maximal fan in 4-factor projection.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.3
• /
• pp.1-15
• /
• 1998

In this paper, we propose a partial scan test method which can be applied to large VLSI designs. In this method, it is not necessary to hold neither scanned nor unscanned flip-flops during scan in, test application,or scan out. This test method requires almost identical design for testability modification and test wave form when compared to the full scan test method, and the method is applicable to large VLSI chips. The well known FAN algorithm has been modified to devise to sequential ATPG algorithm which is effective for the proposed test method. In addition, a partial scan algorithm which is effective for the proposed test method. In addition, a partial algorithm determined a maximal set of flip-flops which gives high fault coverage when they are unselected. The experimental resutls show that the proposed method allow as large as 20% flip-flops to remain unscanned without much decrease in the full scan fault coverage.