Title: A Short Course at Tamkang University Taipei, Taiwan, R.O.C. March 7-9 2006
1An Application of Coding Theory into Experimental
Design
- A Short Course at Tamkang University Taipei,
Taiwan, R.O.C. March 7-9 2006
Shigeichi Hirasawa
Department of Industrial and Management Systems
Engineering, School of Science and Engineering
, Waseda University
21.Introduction
??
31.1 Abstract
????
????
Experimental Design
Coding Theory
????
Error-Correcting Codes (ECCs)
Orthogonal Arrays (OAs)
close relation
Hamming codes, BCH codes RS codes etc.
??? L8
table of OA L8 etc.
キ relations between OAs and ECCs キ the table of
OAs and Hamming codes キ the table of OAs
allocation
41.2 Outline
1.Introduction 2.Preliminary 3.Relation between
ECCs and OAs 4.Conclusion
??
??
??
52.Preliminary
??
6Experimental Design
?????
72.1 Experimental Design (?????)
2.1.1 Experimental Design
Ex.)
??A
キ Factor A (materials)
A0(A company),A1(B company)
a Ratio of Defective Products
??B
キFactor B (machines)
B0(new),B1(old)
??C
キFactor C (temperatures)
C0(100?),C1(200?)
キHow the level of factors affects a ration of
defective products ? キWhich is the best
combination of levels ?
8????
Complete Array
experiments with all combination of levels
A B C
experiment with A0,B0,C0
??
0 0 0 0 0 1 0 1 0 0 1 1 1 0 0 1 0 1 1
1 0 1 1 1
Experiment ?
?
?
?
?
?
?
?
9????
Orthogonal Array (OA) OA(M, n, s,t) (s2)
????
subset (subspace) of complete array
A B C
011
001
0 0 0 0 1 1 1 0 1 1 1 0
Experiment ?
101
111
?
?
000
010
?
100
110
??
strength t2
every 2 columns contains each 2-tuple exactly
same times as row
10????
Orthogonal Array (OA) OA(M, n, s,t) (s2)
????
subset (subspace) of complete array
A B C
011
001
0 0 0 0 1 1 1 0 1 1 1 0
Experiment ?
101
111
?
?
000
010
?
100
110
??
strength t2
every 2 columns contains each 2-tuple exactly
same times as row
11????
Orthogonal Array (OA) OA(M, n, s,t) (s2)
????
subset (subspace) of complete array
A B C
011
001
0 0 0 0 1 1 1 0 1 1 1 0
Experiment ?
101
111
?
?
000
010
?
100
110
??
strength t2
every 2 columns contains each 2-tuple exactly
same times as row
12????
Orthogonal Array (OA) OA(M, n, s,t) (s2)
????
subset (subspace) of complete array
A B C
011
001
0 0 0 0 1 1 1 0 1 1 1 0
Experiment ?
101
111
?
?
000
010
?
100
110
??
strength t2
every 2 columns contains each 2-tuple exactly
same times as row
132.1.2 Construction Problem of OAs
???
the number of factors n3
Parameters of OAs
A B C
???
キthe number of factors n
?
0 0 0
????
?
0 1 1
キthe number of runs M
????
the number of runs M4
?
1 0 1
??
キstrength t2t
trade off
?
1 1 0
this can treat t-th order interaction effect
??
strength t2
Construction problem of OAs is to construct OAs
with as few as possible number of runs, given the
number of factors and strength (n,t ? min M)
142.1.3 Generator Matrix (????)
Generator Matrix of an OA G
Ex.)
orthogonal array 000 , 011 , 101 , 110
A
B
C
A
B
C
0 1 1
(?,?,?) (?,?)
1 0 1
OA
each k-tuple (k2) based on0,12 2kM
generator matrix G
To construct OAs is to construct generator matrix
152.1.3 Generator Matrix (????)
Generator Matrix of an OA G
Ex.)
orthogonal array 000 , 011 , 101 , 110
A
B
C
A
B
C
0 1 1
( 0, 0, 0 ) ( 0,0 )
1 0 1
OA
each k-tuple (k2) based on0,12 2kM
generator matrix G
To construct OAs is to construct generator matrix
162.1.3 Generator Matrix (????)
Generator Matrix of an OA G
Ex.)
orthogonal array 000 , 011 , 101 , 110
A
B
C
A
B
C
0 1 1
( 0, 1, 1 ) ( 1,0 )
1 0 1
OA
each k-tuple (k2) based on0,12 2kM
generator matrix G
To construct OAs is to construct generator matrix
172.1.3 Generator Matrix (????)
Generator Matrix of an OA G
Ex.)
orthogonal array 000 , 011 , 101 , 110
A
B
C
A
B
C
0 1 1
( 1, 0, 1 ) ( 0,1 )
1 0 1
OA
each k-tuple (k2) based on0,12 2kM
generator matrix G
To construct OAs is to construct generator matrix
182.1.3 Generator Matrix (????)
Generator Matrix of an OA G
Ex.)
orthogonal array 000 , 011 , 101 , 110
A
B
C
A
B
C
0 1 1
( 1, 1, 0 ) ( 1,1 )
1 0 1
OA
each k-tuple (k2) based on0,12 2kM
generator matrix G
To construct OAs is to construct generator matrix
19Parameters of OAs and Generator Matrix G
Ex.)
orthogonal arrays 000 , 011 , 101 , 110
the number of factors n3
3
キthe number of factors n3
0 1 1
G
キthe number of runs M4
2
1 0 1
the number of runs M22
キstrength t2
any 2 columns are linearly independent
strength t2
20Parameters of OAs and Generator Matrix
Ex.)
orthogonal arrays 000 , 011 , 101 , 110
the number of factors n3
3
キthe number of factors n3
0 1 1
G
キthe number of runs M4
2
1 0 1
the number of runs M22
キstrength t2
any 2 columns are linearly independent
0
1
0
strength t2
?
1
0
0
21Parameters of OAs and Generator Matrix
Ex.)
orthogonal arrays 000 , 011 , 101 , 110
the number of factors n3
3
キthe number of factors n3
0 1 1
G
キthe number of runs M4
2
1 0 1
the number of runs M22
キstrength t2
any 2 columns are linearly independent
0
1
0
strength t2
?
1
1
0
22Parameters of OAs and Generator Matrix
Ex.)
orthogonal arrays 000 , 011 , 101 , 110
the number of factors n3
3
キthe number of factors n3
0 1 1
G
キthe number of runs M4
2
1 0 1
the number of runs M22
キstrength t2
any 2 columns are linearly independent
1
1
0
strength t2
?
0
1
0
23OAs and ECCs HSS