80 2002 8 6 4 3 5 15 [2] 21 5 13 7 5 3 2 1 1998 GIP [3] 109 55 1 5-10 2 15 20 16 13 [1] 75
3 20 4R&D =0.0411 10 40 Nissan 2006 200-300 2003 12 2003 3 80 4 3 10%-15% [4] QC SED OLAP 1993 30 [5] [6] 20 WSGorsset(1876-1937) RAFisher(1890-1962) Fisher (SED) Tschuprow 1923 JSNeyman 1934 76
20 50 Deming SED Genichi Taguchi SED (TQC) 80 80 1982 2 25 8 1981 12 30 1 3 8 1982 Taguchi 3 [7] 3 60 TQC [6] [8] (QC) TQC [9] [10] 1 ( ) ( ) 77
( 2.1) 2 X ( ) X 2.1 + = (1) (2) 3 78
X X 2.2 3 2.3 2.2 X X 0-1 0-1 ( 2.3) 79
a S 2 S Cv =S/a/ 4 (Process CapabolityPC) PCI X 2.3 ( ) PC PC 3 6 PC6 2.3 6 1-0.9973=0.27% SED [11] [12] 70 80 1 2 Ln(qs) Un(qs) 2 3 1 2 s 80
1 2 q 3 4 3 n n=q 3 4 5 f i X i Y j 6 5 Y j = FX 1 X i X S X q1 1 X qi i X qs S q i i 1 7 X q1 1 X qi i X qs S n+1 8 OLAP IT 2.1 2.1 IT 2001 12 Data Mining DM data set data sets data sets IT IT data bases mining 81
Hand et al. (2000) " Data mining is the process of seeking interesting or valuable information in large data bases" OLAP DM DM Simon OLAP RDB (DCE) [13] [14] 1978 5 10 50-2800 961 31 82
94 1994 10 1995 7 1999 10 2001 5 2002 5 5 20 1 2 3 U 7 7 3 A B ml h C C F A 1 1.0 10 0.5 1 1 2 3 2 1.4 13 1.0 2 2 4 6 3 1.8 16 1.5 3 3 6 2 4 2.2 19 2.0 4 4 1 5 5 2.6 22 2.5 5 5 3 1 6 3.0 25 3.0 6 6 5 4 7 3.4 28 3.5 7 7 7 7 4 A B ml 3.4 C h Y % 1 1.0 13 1.5 0.330 2 1.4 19 3.0 0.366 3 1.8 25 1.0 0.294 4 2.2 10 2.6 0.476 5 2.6 16 0.5 0.209 6 3.0 22 2.0 0.451 7 3.4 28 3.5 0.482 A 1999 SED UD SED UD 3 7 U 7 7 3 [15] P1830 34 5 6 Y = SED 3.4 Y = 0.06689 + 0.24X 3 + 0.0284X 1 X 3-0.0464X 2 2 3 + 0.002578X 2 X 3 + 0.000073X 2 B C 1 1.0 13 ml 1.5 h 2 1.4 19 ml 3.0 h 3 1.8 25 ml 1.0 h 4 2.2 10 ml 2.6 h 5 2.6 16 ml 0.5 h 6 3.0 22 ml 2.0 h 7 3.4 28 ml 3.5 h 0.06232 + 0.251C+ 0.0235AC - 0.06C 2 7 A=3.4 B=19 ml C=2.7575 h Y = 51.84 % 8 83
X 3 X 1 X 3 X 2 3 X 1 X 3 3.4 X 1 X 3 X 2 3 3 X 1 = 3.4 X 2 = 10 X 3 = 3.3464 Y = 59.39 % 7 UD UD 2002 8 15 18 3 1 3 UD UD 2002 7 30-8 1 50 2 2002 5 22-25 200 5 23 1 2002 8 8 UD 70 [16] UD [16] 84
UD UD CIMS UD QC UD UD 1 http://ust40.html.533.net/ xyzok21@sohu.com http://xyzok.126.c 2 http://www.math.hkbu.edu.hk/uniformdesign/society/main.htm 85
Scientific experiments are of essential importance in people's surviving and exploring of nature. Experiments are performed almost everywhere, usually for the purpose of discovering something about a particular process and for the development of a new process. It is usually expected that the experiment could result in increase of the yield, improvement in the quality, reduction of the development time or reduction of the overall costs. The Uniform design is another such efficient fractional factorial design. It was proposed by Professor Fang Kai-Tai and Professor Wang Yuan in 1980. It has been successfully used in various fields such as chemistry and chemical engineering, pharmaceutics, quality engineering, system engineering, survey design, computer sciences and natural sciences. The uniform design has been recoginized as an important space-filling design by the internatinal community. The space-filling design has played an important role in large system engineering. The uniform design is also one of the robust designs. 3 http://www.zhangying.com/frameset.htm. 1946.2 89 :, ISO-9000, " [1] 2003.8 2003 8 16 V8.B [2] 2003.8 2003 8 14 [3] 2003 2002/2003 [4] 2003.10 [5] 30 (1993) 10.25,10.27,11.17,11.19 [6] 1997.3 ---- () P64-90 86
[7] 1985.3 [8] 1975.9 [9] Genichi Taguchi, Yuin Wu (1979) Introduction to off-line Quality Control,Central Japan Quality Assoiation. : ( 1979 QC) [10] 1999 9 [11] 1978 [12] 1981 [13] Fang,K.T. Wang,Y. 1993Number-Theoretic Method in Statistics, Chapman and Hall, London: 1996 [14] 2001 [15] 1994 6 [16] 2002.8 87