Professional Documents
Culture Documents
PART I
Andrew Kusiak Intelligent Systems Laboratory 2139 Seamans Center The University of Iowa Iowa City, Iowa 52242 - 1527 Tel: 319 - 335 5934 Fax: 319 - 335 5669 andrew-kusiak@uiowa.edu http://www.icaen.uiowa.edu/~ankusiak
The University of Iowa Intelligent Systems Laboratory
Contents
INTRODUCTION CLUSTER ANALYSIS METHOD BRANCHING ALGORITHMS ASSIGNMENT OF PARTS TO THE EXISTING MACHINE CELLS SUMMARY
The University of Iowa Intelligent Systems Laboratory
Grouping Parts
Grouped Parts
PF - 1 PF - 2
PF - 3
PF - 4
Grouping Routes
Ungrouped Parts and Machines
P1 PF - 1 M1 P3 M4 P1 P2 M6 M2 M5 P3 P2 M3 P5 M6 P4 P5 M1
Group Technology
Reduced production lead time
MC - 1 M2
P4
PF - 2 M3 M4
MC - 2
M5
Parts
Machines
Parts Families
Machines Cells
Reduced work-in-process Reduced labor Reduced tooling Reduced rework and scrap materials Reduced setup time Reduced order time delivery Improved human relations Reduced paper work
Intelligent Systems Laboratory
Page 1
1
Visual Method
Two Basic Methods of Group Technology:
PF - 1 PF - 2
Classification
- Visual method - Coding method
PF - 3 PF - 4
Cluster analysis
The University of Iowa Intelligent Systems Laboratory
Grouped parts
Intelligent Systems Laboratory
Sample monocode 3 1 2
Digit 2
No center hole
?
Question:
Is the classification and coding method exact or ambiguous?
Ambiguous!
The University of Iowa Intelligent Systems Laboratory
Page 2
2
Questions:
Implementation of GT:
Physical Machine Layout
Parts P1 Machines MC - 1 PF - 1 M1 P3 M4 P1 P2 M6 M2 M5 P3 P2 M3 P5 Ungrouped parts and machines Grouped parts and machines
Intelligent Systems Laboratory
What is the main difference between the classification and coding approach and the cluster analysis approach? Which of the two GT approaches requires information at an earlier phase of the product development cycle?
The University of Iowa Intelligent Systems Laboratory
M1
M2
P4
PF - 2 M3 P5 M4
MC - 2
M6
P4
M5
Implementation of GT:
Logical Machine Layout
Parts P1 PF - 1 M1 P3 M4 P1 P2 M2 P4 M2 M5 P5 P2 M3 P5 M6 P3 P2 M3 M6 PF - 2 M5 M1 M4 Machines MC - 1 MC - 2
PF-2 2 4 5
Matrix Formulation
1 Part number 1 1 Machine 2 number 3 4 1 1 1 1 2 1 1 1
MC-2 3 1 1 PF -1 PF-2 2 4 5
2 1
4 1
5 1 2 MC-1 4
1 1 1
3 1 1
2 3 4
1 1 1
1 1 MC-2 1
4 1
5 1
2 MC-1 4 1
1 1 1
3 1 1
1 3
1 1
1 1
2 1 1
5 1
1 1
Bottleneck part
Page 3
3
Bottleneck part 5
1 2 Machine 3 number 4 5
1 1 1
1 1 1
1 1 1
1 1 1
bottleneck machine 3
1 MC-1 { 2 3(1) 3(2) MC-2 { 4 5
PF-1 1 1 1 1 2 1 1 1 1 1 1 1 3 4
PF-2 5 6
The bottleneck part 5 prevents the decomposition of a machine-part incidence matrix into disjoint submatrices
The University of Iowa Intelligent Systems Laboratory
1 1 1
1 1 1
Question:
?
1 2 1 1 1 1 1 1 1 3 4 5 1 1 1
Solving the GT problem represented with a binary matrix: Similarity coefficient methods Sorting based algorithms Bond energy algorithm Cost-based method Cluster identification algorithm Extended cluster identification algorithm
The University of Iowa Intelligent Systems Laboratory
Page 4
4
Example
Incidence matrix
1 2 3 *
Part
1 2 3 4 5 6 7 8 * * * * * * * * * * * * * *
Machine 4 *
5 6 7
The University of Iowa
1 2 3 4 5 6 7
1 2 3 4 5 6 7 8 * * * * * * * * * * * * * * * *
Intelligent Systems Laboratory
Next Step
1
1 2 3 4 5 6 7 8 * * * * * * * * * * * * * * * *
1 4 6 7 2 3 4 6 * * * * * * *
2 3 4 5 6 7
Resultant Matrix
Iteration 2
1 4 6 7 2 3 4 6 * * * * v1 v6
4 7 * * * v4 v7
* * *
h2 h4
1 4 6 7 2 3 4 6 * * * * *
Iteration 3
* *
3 6
h3 h6
Page 5
5
CI Algorithm Repeated
(For notes follow up)
Step 0. Set iteration number k = 1. Step 1. Row 1 of matrix (6) is selected and horizontal line h1 is drawn. Step 2. Three vertical lines v2, v3, and v5 are drawn for each single-crossed entry 1.
1 1 2 3 4 5 6 7 1 1 1 1 1 1 v
2
Step 3. Two horizontal lines h5 and h7 are drawn through the single-crossed entries of last matrix
1 1 2 3 4 5 6 7 1 1 1 1 1 1 v
1
2
2 1
3 1
5 1
6 1
8 h
1
1 1 1 1 1 1
5
2 1
3 1
5 1
6 1
8 h
1
1 v
3 1 1
3
v
4 5 1
1 1 1 1 1 1
5
2 1
6 1
8 h1
1 v
3
1 2 3 4 5 6 7
1 1 1 1 1 v2 1 v3 1 v5 1
1 1 1 1 h5 h7
Step 4. Since the entries (5, 8) and (7, 8) are single-crossed, the vertical line v8 is drawn.
1 1 2 3 4 5 6 7 1 1 1 1 1 1 v
2
2 1
3 1
5 1
6 1
8 h
1
1 1 1 1 v
5
1 1 v
8
h h
5 7
1 v
3
In the second iteration (k = 2), Steps 1 through 4 are performed for matrix above. This iteration results in the following incidence matrix:
1 4 1 1 1 v
1
6 1
7 h 1 h
2 4
As there are no more single-crossed entries, all double-crossed entries 1 form: Machine cell MC-1 = {1, 5, 7}, and Part family PF-1 = {2, 3, 5, 8}.
The University of Iowa Intelligent Systems Laboratory
2 3 4 6
1 v
Page 6
6
In the third iteration (k = 3), removing double-crossed entries from matrix below
4 3 6 1 1 v
4
7 1 v h h3
6 7
From this matrix MC-3 = {3, 6} and PF-3 = {4, 7} are obtained. The final clustering result with three machine cells and part PF-1 PF-2 PF-3 families:
2 1 MC-1 { 5 7 2 MC-2 { 4 3 MC-3 { 6 The University of Iowa 1 1 3 1 1 1 5 1 1 1 8 1 1 1 1 1 1 1 1 1 1 6 4 7
Parts: 1, 6
Extended CI Algorithm
Step 0. Set iteration number k = 1. (k) Step 1. Select those machines (rows of matrix [aij] ) that based on the user's expertise, are potential candidates for inclusion in machine cell MC-k. Draw a horizontal line hi through each row of matrix [aij](k) corresponding to these machines. In the absence of the user's expertise any machine can be selected. (k) Step 2. For each column in [aij] corresponding to entry 1, single-crossed by any of the horizontal lines hi , draw a vertical line vj . (k) Step 3. For each row in [aij] corresponding to the entry 1, single-crossed by the vertical line vj, drawn in Step 2, draw a horizontal line hi . Based on the machines corresponding to all the horizontal lines drawn in Step 1 and Step 3, a temporary machine cell MC-k , is formed.
The University of Iowa Intelligent Systems Laboratory
If the user's expertise indicates that some of the machines cannot be included in the temporary machine cell MC-k, (k) erase the corresponding horizontal lines in the matrix [aij]. Removal of these horizontal lines results in machine cell MC'-k . Delete from matrix [aij] (k )parts (columns) that are to be manufactured on at least one of the machines already included in MC-k. Place these parts on the list of parts to be manufactured in a functional machining facility. Draw a vertical line vj through each single-crossed entry 1 in [aij](k) which does not involve any other machines than those included in MC-k .
(k)
Step 4. For all the double-crossed entries 1 in [aij] , form a machine cell MC-k and a part family PF-k
The University of Iowa Intelligent Systems Laboratory
Step 5. Transform the incidence matrix [aij] into [aij] (k+1) by removing all the rows and columns included in MC-k and PF-k, respectively. Step 6. If matrix [aij] (k +1)= 0 (where 0 denotes a matrix with all elements equal to zero), stop; otherwise set k = k + 1 and go to step 1.
(k)
2 1
3 1
5 1
7 1
10
11 1 Machine number
1 1 1 1 1 1 1 1 1 1 1 1 1
Page 7
7
Step 0. Set iteration number k = 1. Step 1. Since user's expertise indicates that machines 1 and 4 should be included on machine cell MC-1, two horizontal lines h1 and h 4 are drawn.
Part number 1 1 2 3 4 5 6 7 1 1 1 1
The University of Iowa
2 1
3 1
5 1
7 1
10
11 h1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 h4
Step 2. For columns 1, 2, 3, 6, and 7 crossed by the horizontal lines h 1 and h 4, five vertical lines, v1, v2, v3, v6, and v7 are drawn.
Part number 1 1 2 3 4 5 6 7 1 1 1 1 v1 v2 v3 1 1 1 1 1 v6 1 v7 2 1 3 1 1 1 1 1 1 1 1 1 h4 4 5 6 7 1 1 8 9 10 11 h1
Step 3. Three horizontal lines, h2, h6, and h7 are drawn through rows 2, 5, and 7 corresponding of the single-crossed elements 1.
1 1 2 3 4 5 6 7 1 1 1 1 v1 v2 v3 1 1 1 1 1 v6 1 v7
Intelligent Systems Laboratory
2 1
3 1
Part number 5 6 7 1 1
10
11 1 h1 h2 h4
1 1 1 1 1 1 1 h6 h7
A temporary machine cell MC'-1 with machines {1, 2, 4, 6, 7} is formed. Machines 2 and 6 are excluded from in MC'-1 as they include less double-crossed (committed) elements than machine 7. The horizontal lines h2 and h6 are erased.
2 1 2 3 4 5 6 7 1
3 1
5 1
7 1
10 11 h1 1 1 h4
1 1 1 v2 v3
1 1 1 1 v6 1 v7 1
h7
Intelligent Systems Laboratory
M = 1, 4, 7
The University of Iowa Intelligent Systems Laboratory
Page 8
8
Step 4. The double-crossed entries 1 of matrix indicate: Machine cell MC-1 = {1, 4, 7} and Part family PF-1 = {2, 3, 6, 7} Step 5. Matrix is transformed into the matrix below.
Step 6. Set k = k + 1 = 2 and go to Step 1. The second iteration (k = 2) results in: Machine cell MC-2 = {2, 3, 5, 6} and Part family PF-2 = {5, 8, 10, 11}
5 2 3 5 6 1
10 1
11 1 h2 h3 h5 h6
5 2 3 5 6
The University of Iowa
10 1
11 1
1 1 1 1 1
1 v5
1 1 v8
1 v10 v11
Page 9
9