Group Technology

Rank Order Clustering, To-From Analysis,

Cells
Dr David J. Newlands
IESEG
2016

Product Families
Group Technology (GT) is commonly used to
identify product families based on similarities in
design or processing
GT helps reduce unnecessary duplication in
product design
GT identifies product families with similar
processing requirements
These families can be manufactured in welldefined manufacturing cells

Group Technology Scheme

(b) Grouped Cylindrical Parts (families of parts)
(a) Ungrouped Parts
Grooved

Slotted

Threaded

Drilled

Machined

Figure 5.10

An Example of a Dendogram
A

100%
90%

60%

30%

The Math only divides into two at any one %.

4

Represents a processing unit P.U.

26

Represents a process route between two P.U.s that

services 26 different components i.e. frequency 26

Factory Flow Analysis

SIMPLE

COMBINING

COMPLEX

2+3

COMBINING

234
5

Factory Flow Analysis Redeployment of Plant

SIMPLE

1
2 1

COMPLEX

1
1
2

234
1

Avoid Looping and Backflows

NO LOOPING
1

NO BACKFLOW
2

FACTORY FLOW ANALYSIS - Example

Route A

Saw S/C

S/C

S/C

S/C

Saw M/C

HT

M/C

M/C

M/C

Plate M/C

M/C

Assy H/T

H/T

Assy Plate M/C

HT

M/C

M/C

M/C

Assy Plate

Assy
25

S/C

S/C

M/C

Assy Assy HT

S/C

Assy M/C

Assy Assy

Assy

Assy
35

30

10

10

10

16

Basic Flow Chart Answer

142
35

Saw

Store

25
2

10

105

4
90

HT

M/C

65
4

S/C

76
1
10
20

90
1
Assy

142

10

16
30

Simplified Flow Chart

142
Store

Assy

11

Simplified Flow Chart

35

Saw

Combine HT & M/C

Relocate subass G in M/C P.U.
Sub contract J 1st M/C op.

Store
1

35

HT
M/C
Sub Ass

106

S/C

80

10
20

96

142

16
30

Assy
142

12

Spaghetti Drawing-Before

Source: http://www.seaonline.org/docs/Regent.ppt

Spaghetti Drawing-After

14
Source: http://www.seaonline.org/docs/Regent.ppt

M i e s te n
p u k u ti l a

PC

Ul o s k yn ti

Si s n k y n ti

ilm a s t o in t i-

k a n a v a

PPP

M AT ER.

SY T T

T u pa k k a

Vlisein

FA X
PRINTER

COPY

PPP

NEUKKARI

pos tipak e tit

s hk k aappi

SHKK.

.00 cm

Kryimuri

PESU

PPP

KATISKA

NEUKKARI

M E M M ER T

No stin

1.5 m

.00 cm

s hk k aappi

Ty p e n ke h i ti n

Na i s te n
p u k u ti l a

Material
replenishme
nt
M i kro s k .

DC-CONNECTION

Raw
material
storage

SAT

.00 cm

PESUP.
P uhdasvesiasema

Va a te s ily ty s

MPM

.00 cm

JK

Receiving
and
inspection
JK

.00 cm

.00 cm

.00 cm

.00 cm

.00 cm

Nokia - Salo Operations Material Flow

Recyclable and waste
Example
Incoming
material

materials
Outgoing Products

.00 cm

Shipping to
customers

3.9 m

4.7 m

Finished
products
storage

Or
packaging in
DC

X- ray
shkkaappi

On-line
packaging

Component / Machine Chart

Part
Machine
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P

E
F
H

1
1
1
1
1
1
1

1
1
1

1
1

1
1
1
1

1
1

1
1

1
1

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
1
1
2
1
1
1
1 1
1
1
8
1
1 1 1
5
1
1
1
1
1
1
7
1
1 1 1
1
1
1
1
1
1
1 13
1 1 1
1
1
1
1 1 1
1
1 1
1 1 19
1
1
3
1
1 1
1
1 1 1
1 1
1 1
1
1 1
1
1 20
1
1
1
1
1 1
1
1
10
1 1
1 1
1
1
7
1
1
1
1
1
6
1
1
1
1
1
5
1
2
1
1
4
1
1
1
1
1
1 7
1
1
1
1 1
1
8

20% OF MACHINES WITH HIGHEST FREQUENCY IN PARTS PER M/C. THESE M/Cs ARE TAKEN OUT THEN READMITTED.
1
1 1
1 1 1
1
1
1
1
1
1
1 1 1
1 1 1
1
1
1
1 1 1
1
1 1
1 1
1 1
1
1 1 1
1 1
1 1
1
1

16

1
1
1

1
1

1 13
1 19
1 20

Division into Family Groups

Part
Machine
J
F1
G
H1
I
B
P
F2
H2
A

1
1
1
1
1

13 39 25 12 31 26 42 37 2 32 38 10 40 28 18 4 27 24 3 20 30 11 22 17 7 35 6 34 36 19 23 14 43 5 9 21 41 15 29 8 33 16
1 1 1 1 1 1
1 1
1
1
1
1 1
1
1
1
1

1
1
1
1
1
1 1

1
1
1
1
1

1 1 1 1 1 1 1
1 1 1 1 1
1 1 1
1
1
1
1
1

K
L
H3
M
C
F3
N
E
D
F4
O
H4

exception
1 1 1 1 1
1 1
1 1 1
1 1 1 1
1
1
1

exception

exception
1 1 1
1 1
1 1
1 1
1
1 1
1
1
1
1
1

1 1 1 1 1 1 1 1 1 1 1 1
1 1
1 1 1
1
1 1 1
1 1
1 1 1
1 1
1
1
1
1 1 1 1
1

There is no absolute answer, only an optimum!

Modify the features to fit standard cells or modify the cell (re-plan) or subcontract.
If utilisation is low, consider mixing cells.

Rank Order Clustering Example /

Question

MACHINES

COMPONENTS

18

A
B
C
E
F
G
H
I

1 2 34
X X
XX
XX
X X
X
X
X
X

5 6 78
X X
X X
X X
X X
X X
X
X X
X

9 10
X

X
X X
X
X

RANK ORDER CLUSTERING WORKSHEET

MACHINES

COMPONENTS

MACHINES

MACHINES

COMPONENTS

19

COMPONENTS

MACHINES

MACHINES

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
XX
X X X
XX
X XX
XXX X XX
XX X
X
X
X
X
XX
X
X
X

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

RANK ORDER CLUSTERING WORKSHEET

COMPONENTS

1 2 3 4 5 6 7 8 9 10
A
X X X X
128
B XX
X X
X 64

MACHINES

XX

E
F

G X
H

X
X X X

X X

X
X

I X

X X

16
8
4

X
X X

101 96 144154 152 101 154 98 52

20

32

X X

X
81

RANK ORDER CLUSTERING WORKSHEET

COMPONENTS

MACHINES

COMPONENTS

MACHINES
64
32
16
8
4
2
1

512
256
128

4 7 5 3 1 6 8 2 10 9

21

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

64
32
16
8
4
2
1

101 96 144 154 152 101 154 98 52 81

MACHINES

4 7 5 3 1 6 8 2 10 9
AXXXX
B
X XX X X
C
X XX X X
E XXXX
XX
F XXX
G
XX
X
HXX
X
I
XX
X
512
256
128

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
128
XX
X X X 64
32
XX
X XX
X X X X X X 16
8
XX X
4
X
X
X
XX
X
2
X
X
X 1

MACHINES

MACHINES

COMPONENTS

22

9
960
62

61

963
896

64
32
16
8
4
2
1

256
128

4 7 5 3 1 6 8 2 10
A X X X X
X X XX X
B
X X XX
C
E
X
X X X X
F X X X
G
X X
X
H X X
I
X X
X
512

MACHINES

COMPONENTS

49
776
50

RANK ORDER CLUSTERING WORKSHEET

MACHINES

4 7 5 3 6 1 8 2 10 9
E XXXX
X X 128
64
AXXXX
32
F XXX
16
HXX
X
8
B
XXXXX
C
XXXX X 4
2
I
XX
X
XX
X 1
G

COMPONENTS

MACHINES

MACHINES

COMPONENTS

4 7 5 3 61 1 8 2 10 9
AXXXX
B
X XX X X
C
X XX X X
E XXXX
XX
F XXX
G
XX
X
HXX
X
I
XX
X

23

COMPONENTS

MACHINES

MACHINES

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
XX
X X X
XX
X XX
XXX X XX
XX X
X
X
X
XX
X
X
X
X

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

RANK ORDER CLUSTERING WORKSHEET

MACHINES

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
X XXX X
B
C
XXXX X
I
X
XX
X XX
G

4 7 5 3 6 1 8 2 10 9
E XXXX
X X 128
64
AXXXX
32
F XXX
16
HXX
X
8
B
XXXXX
C
XXXX X 4
2
I
XX
X
XX
X 1
G

COMPONENTS

MACHINES

MACHINES

COMPONENTS

4 7 5 3 61 1 8 2 10 9
AXXXX
B
X XX X X
C
X XX X X
E XXXX
XX
F XXX
G
XX
X
HXX
X
I
XX
X

24

COMPONENTS

MACHINES

MACHINES

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
XX
X X X
XX
X XX
XXX X XX
XX X
X
X
X
XX
X
X
X
X

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

RANK ORDER CLUSTERING WORKSHEET

25

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
B
X XXXX
I
XX
X
XXXXX
C
X XX
G

4 7 5 3 6 1 8 2 10 9
E XXXX
X X 128
64
AXXXX
32
F XXX
16
HXX
X
8
B
XXXXX
C
XXXX X 4
2
I
XX
X
XX
X 1
G
COMPONENTS

MACHINES

COMPONENTS

MACHINES

MACHINES

COMPONENTS

64
32
16
8
4
2
1

101 96 144 154 152 101 154 98 52 81

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
X XXX X
B
C
XXXX X
I
X
XX
X XX
G

4 7 5 3 61 1 8 2 10 9
AXXXX
B
X XX X X
C
X XX X X
E XXXX
XX
F XXX
G
XX
X
HXX
X
I
XX
X
512
256
128

MACHINES

MACHINES

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
128
XX
X X X 64
32
XX
X XX
X X X X X X 16
8
XX X
4
X
X
X
XX
X
2
X
X
X 1

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
B
X XXXX
I
XX
X
XXXXX
C
X XX
G

RANK ORDER CLUSTERING WORKSHEET

64
32
16
8
4
2
1

512
256
128

26

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
X XXXX
B
I
XX
X
C
XXXXX
X XX
G

4 7 5 3 6 1 8 2 10 9
E XXXX
X X 128
64
AXXXX
32
F XXX
16
HXX
X
8
B
XXXXX
C
XXXX X 4
2
I
XX
X
XX
X 1
G
COMPONENTS

MACHINES

COMPONENTS

MACHINES

MACHINES

COMPONENTS

64
32
16
8
4
2
1

101 96 144 154 152 101 154 98 52 81

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
X XXX X
B
C
XXXX X
I
X
XX
X XX
G

4 7 5 3 61 1 8 2 10 9
AXXXX
B
X XX X X
C
X XX X X
E XXXX
XX
F XXX
G
XX
X
HXX
X
I
XX
X
512
256
128

MACHINES

MACHINES

A
B
C
E
F
G
H
I

1 2 3 4 5 6 7 8 9 10
XXX X
128
XX
X X X 64
32
XX
X XX
X X X X X X 16
8
XX X
4
X
X
X
XX
X
2
X
X
X 1

COMPONENTS

MACHINES

COMPONENTS

COMPONENTS

4 7 5 3 10 9 8 6 1 2
E XXXXXX
AXXXX
F XXX
HXX
X
X XXXX
B
I
XX
X
C
XXXXX
X XX
G

Dealing with Exceptions

Change the component specification to eliminate the process

Redesign products to eliminate exception processes
Modify the manufacturing sequence to push to the beginning or
end activity
Subcontract the individual activity particularly first and last
activities
Purchase instead of making the entire component particularly
older designs (then focus in-house on products with high value add
and less than 2 year old designs, focus on cost elimination at
supplier)
Co-locate cells to share the process, particularly for low utilised
machines.
Consider purchasing / leasing a new machine to deal with
exceptions where co-location isnt possible
Re-routing exceptional operations to machines inside the group if
capacity allows
Create a further division of machines between groups
Consider super-cells in a jobbing environment join two or more
cells into a crazy line
27

To From Analysis

Line Layout
PRODUCT
X

12 Mfg Organization

COVENTRY UNIVERSITY

Layout Evaluation
Product Volume and Route data
Product
Volume
Route
1
10
A-B-C-D-E
2
15
A-B-D-E
3
6
A-C-B-D
Volume Matrix
From\ to A B
A
25
B
C
6
D
E

C
6
10
-

D
21
10
-

25
-

This starts the process of layout evaluation

Data is collected about the products, or components,

to be produced in the work area, the volume (per week)
and the 'process route' ( set of processes in sequence)
The top table lists this data for each product
The bottom table converts this data into a volume matrix
This shows the total volume/week on each transport
route (move between each pair of processes/facilities)

This gives an indication of which facilites should be close

together (because there is a high flow between them)
The volume matrix relates to products NOT to a layout

Volume Distance Analysis

Distance Matrix

From\To
A
B
C
D
E

A
10
15
20
12

B
10
10
15
18

C
15
10
10
12

D
20
15
10
10

E
12
18
12
10
-

Volume - Distance Matrix

From\To A
A
B
C
D
E

B
C D E
250 90
100 315
60 100
250
-

Total

1165

This continues the process of layout evaluation

For a current or proposed layout, the distance between

each pair of facilities can be measured and listed in a
distance matrix. This relates to layout NOT to products.

The volume from the volume matrix is multiplied by the

distance from the distance matrix (each cell separately)
to form the volume-distance matrix.
The Volume-distance total is calculated ( = 1165 here)
This value measures the amount of internal movement
occurring for these products for a given layout.
A revised layout would be more efficient if it had a lower
volume-distance total.
The high values in the vol.-dist. matrix can be reduced by
locating closer facilities with a high volume between them

Determinging an Efficient Layout

Example
Determing forward flow start-finish (From-To)
Determine how many
Determine how far
Rank more from to more to.
Determine Efficiency by identifying inefficiency
first
Set requirement that all future products are
developed/designed to fit the new optimum
sequence to increase future forward efficiency

To-From Data to Optimise Flow and

Minimise Back flow or Looping
To

From
1
2
3
4
5
6

1
0
12
5
20
14
0

2
0
0
25
10
0
12

3
15
0
0
0
12
8

4
5
18
3
0
10
8

5
0
6
8
22
0
25

6
10 = 30
0
7
16
8
0

=51

Determine the preferred layout order of machines that minimises back flow.
Hence, express as a percentage of all work movements, the extent of back flow
for the idealised layout.
35

To-From Ratio and Optimum Sequence

MC

To

From Ratio

1
2
3
4
5
6

Optimal Sequence =
36

Optimum Flow
Ratio
Sequence

To-From Ratio and Optimum Sequence

MC

1
2
3
4
5
6

To From Ratio
51
30 1.70
47
36 1.31
35
48 0.73
44
68 0.65
61
44 1.39
41
53 0.77

Optimal Sequence = 4
37

Ratio
0.65
0.73
0.77
1.31
1.39
1.70

Optimum Flow
Sequence
4
3
6
2
5
1

5 1

Determine Back flow / Looping as a %

To
From
1
2
3
4
5
6
1
0
0 15
5
0 10
2 12
0
0
18
6
0
3
5 25
0
3
8
7
4 20 10
0
0
22 16
5 14
0 12
10
0
8
6
0 12
8
8
25
0
Total number of movements = 279

Optimal Sequence = 4
All backflows from 1 = , ,
All backflows from 5 = , ,
All backflows from 2 =
All backflows from 6 = ,
All backflows from 3 =

2 5 1

Total backflow =
Backflow % =
38

=
279

. %

Identify back flow away from the customer

4 3

2 5 1 4 3

2 5 1

These are okay

4 3

2 5 1

4 3

This is okay

4 3
39

2 5 1

2 5 1

Determine Back flow / Looping as a %

To
From
1
2
3
4
5
6

1
0
12
5
20
14
0

2
0
0
25
10
0
12

3
15
0
0
0
12
8

4
5
18
3
0
10
8

5
0
6
8
22
0
25

6
10
0
7
16
8
0

Total number of movements = 279

All backflows from 1 = 15, 5, 10
All backflows from 5 = 12, 10, 8
All backflows from 2 = 18
All backflows from 6 = 8, 8
All backflows from 3 = 3
40

Total backflow = 97
Backflow % = 97 = 34.77%
279
Forward flow 65 %

Coping with back flow away from the customer

4 3

2 5 1 4 3

2 5 1

These are okay

4 3

2 5 1

4 3

This is okay

4 3
41

2 5 1

2 5 1

If you are walking, you are not working

Use Cells to Minimise Distances Travelled

Forward and Backward Using Cellular Layouts

Backflow

Forward Flows

Backflow is acceptable when volumes are low

Redesign products to eliminate backflow
Create demand then set aside specific cells for specific product families
to eliminate backflow
Outsource low volume products or withdraw them if they do not fit
forward flow
Examine the data and expect
to re-layout plant every six months or when
42
ever there are changes to product designs and marketing mix

Line Balancing

Line Balancing Example

Tasks for assembly of a product (times in std.min.)
Task
1
2
3
4
5
6
7
8
9
Time
1.2 0.9 0.7 0.9 1.5 1.8 0.4 0.4 1.5
2
1

Output = 240/day

6
4

7
8

day = 480 min

TAKT time = 2 min

Workstations for assembly of a product

Tasks
1+3
2+4
6
5+7
Time
1.9
1.8
1.8
1.9

8+9
1.9

This shows the process of line balancing

This applies to a set of tasks on a mfg./assembly line.

The initial durations for each task will be different.
However the time available per operator is the same,
some operators will be underloaded, the line inefficient.
The line can be made efficient by making the task time
for each operator the same ( or nearly the same).
This is known as 'line balancing' or' balancing the line'.
(NOT the same as 'line of balance' in next session)
The tasks will have a logical sequence of dependencies
(which task must be done to enable a task to start)
Tasks are combined within logic constraints so that
the set of task times are about the same

Line Balancing
Grouping of tasks into workstations

min. workstations
= 9.3 / 2 = 4.65
rounded up to 5

This shows the grouping of tasks from the previous OHV

1 & 3 can be combined since 3 can start when 1 is done

2 & 4 can combine and start when 1 is done
6 is large enough to stand on its own
5 & 7 can combine after (2 + 4) and 6 have been done
8 & 9 can combine when everything else has been done

The result is 5 work stations (or operators) each with a

task of about 1.9 mins. This is the 'cycle time' for the line
.
The Capacity of the line = 240 per 8 hour day

Burndown Charts

http://en.wikipedia.org/wiki/File:Burn_down_chart.png

Japanese lean expression

you cant make the monthly target
unless you make the daily target

Slowing? Not speed! Control variability in rate!!!

=> constant rate => predictability => control

http://en.wikipedia.org/wiki/File:SampleBurndownChart.png

Comparison of Set up and Variable Costs

Under Various Lot Sizes
Variable Cost per Unit

Number
of Units
1
10
100
1000
2000

Set Up
Cost
1 000
1 000
1 000
1 000
1 000

10 000

000
000
000
000
000

51

Total
Variable
Cost
10
100
1 000
10 000
20 000

000
000
000
000
000

Total
Cost
1 010
1 100
2 000
11 000
21 000

000
000
000
000
000

Cost Per
Unit
1 010 000
110 000
20 000
11 000
10 500

Logistics Impact on "Return on Investment",

Customer Service
Logistics
Efficiency

Inventory policies,
Service Levels

Sales revenue

Profit

Costs

Inventory

+
Order completion
rate
Invoice accuracy
Purchasing
policies

Accounts
receivable

+
Cash

+
Fixed Assets
Manufacturing &
distribution facilities
Transportation equipment
52

Capital Employed

ROI

 If you have 1 million, and the

machine you want costs 1 million,
what do you do?

53

Potential Solutions

Just buy it
Lease it
Borrow against money in the bank
..

Japanese solution: Buy 10 machines, pay 10% on each

(1M paid). Use the extra capacity to reduce fixed costs.
Do not locate in one location. Make where you sell, buy
where you make. Global Market, make locally.
By stimulating large demand, significant reductions in
variable cost can be achieved via economies of scale at
all preferred suppliers. Reduce the number of
suppliers to ensure they achieve economies of scale.
Select the best, highest quality, fastest delivery, and
most willing to learn and contribute to design of the
process and product improvement schemes.
54

Comparison of Set up and Variable Costs

Under Various Lot Sizes
610
000
000

Variable Cost per Unit

Number
of Units

000
000
000
000
000

Cost Per
Unit
1 010 000
110 000
20 000
11 000
10 500

20 000 1 000 000 120 000 000 121 000 000

6 050

1
10
100
1000
2000

Set Up
Cost
1 000
1 000
1 000
1 000
1 000

Total
Variable
Cost
10
100
1 000
10 000
20 000

000
000
000
000
000

55

000
000
000
000
000

Total
Cost
1 010
1 100
2 000
11 000
21 000

A
Narrow
Kitchen
The point on top of the head traced onto the floor is a straight line.
Everything is within arms reach. Difficulty to get passed.

56

59

Traditional Jobbing Machine Shop Layout

Six Machines, Six Skilled Men

60

Traditional Jobbing Machine Shop WIP

Six Machines, Six Skilled Men

61

Productive Traditional Machine Layout

62

Twelve Semi-Auto/NC Machines, Six Skilled & Semi-skilled Men

Productive Traditional Machine Layout

Batch Manufacture => WIP

63

Twelve Semi-Auto/NC Machines, Six Skilled & Semi-skilled Men

64

65

Eighteen Semi-auto machines, Six Skilled Men

Machine Cell Layout

Kanban
Bin In

Kanban
Bin Out

Kanban
Bin In
66

Kanban
Bin Out

Kanban
Bin In

Cellular Manufacturing
21 Automated NC Machines (Inc Auto Eject) 1, 2, 3, 4, 6, 8 Semi-skilled Men/Cell

Kanban
Bin Out

Flexible Manufacturing Cells

Kanban
Bin In

67

http://ocw.mit.edu/OcwWeb/Sloan-School-of-Management/15-769Operations-StrategySpring2003/CourseHome/index.htm

Machine Tools, Buffers and an AGV

68

69

Traditional Bicycle Plant Layout and Flow

Tube Stock
Storage

Finished
Stores
and
Shipping

Final
Assembly

Tube
Cutting

Component
Parts
Storage

Frame
Storage

Frame
Parts
Storage

Deburr

Frame
Welding

Frame
Painting

70

Source: Womack & Jones (1997:57)

Tube
Miter

Tube
Bending

Frame
Wash

Lean - Cellular Bicycle Plant Layout and Flow

Bend

Miter

Cut

Deburr

Steel bike cell

Tube
Stock
&
Parts

Weld

Wash Paint

Bend

Miter

Deburr

Assemble

Cut

Aluminium bike cell

Weld
Wash Paint

Bend

Miter

Deburr

Assemble

Cut

Ship

Half the space

Half the time
Half the human effort
Half the costs
10% of the stock
Double the sales volume
Almost no customer waiting for orders
Maximum flexibility
Little pipeline liability
Make to order => No risk to profit

Titanium bike cell

Weld
Wash Paint

Assemble

71

Source: Womack & Jones (1997:62)

A Simulation of Cellular Manufacturing

Two Handed Operations,
One Handed Operations,
and
Loading Kanban Bins
Dr David J. Newlands 2004

72

Traditional Shop Floor Layout

Receiving
Inspection

Stores
Stores

Machining

Test
Shipping

Assembly

Fabrication

Complexity
Conflicting Objectives
Protectionist
Barriers to teamwork/problem solving

Long lead-times and Unresponsive

73

Group Technology Shop Floor Layout

Receive
Store
Store
Store

Store

M
M

F
FA

F AT A
M
A

A
A

T
T
T

S
S
S

Grouped by product family

Economies of scope
Common objectives
Immediate feedback / good communications
Short lead-times and Flexible with no, or little backward flow
74

Steps to Achieving Cellular Manufacturing

1
2
3

4
5
6
7
8
9
10
75

Find out the parts that are made and used, and map how the
process brings them together.
Work out the sequence of operations to achieve flow.
Design the layout to minimise movement and prevent buildup of stock within the process. Select equipment to make future
improvements easy to implement.
Plan to produce to demand with minimal batch sizes.
Use automation and visual management to release people from
waiting.
Design mistake-proofing (poka-yoke) into the process.
Balance the times for each operation.
Train people to allow them to move freely between
operations.
Decide performance measures for the cell.
Create a culture of ownership, responsibility and continuous
improvement.

Line Layout

PRODUCT
X

Process / Job Layout

Product X

(24)

76

(10)

Cell Layout
Product Y

(6)

77

78

79

Characteristics

Job shop

Flow shop

Project shop

Continuous
Process

Types of
machines

Flexible; general
purpose

Special purpose;
single functions

General purpose;
mobile

Special purpose

Design of
processes
Set-up time

Functional or
process
Long, variable

Workers

Single functioned,
highly skilled (1
man to 1 machine)
Large inventory to
provide for large
variety
Small to medium

Product flow layout Project or fixed

position layout
Variable
Long
Single function
One function;
skilled; 1 man to
lower skilled
1 machine
Large to provide
Variable, usually
buffer storage
raw materials

Inventories

Lot sizes

Product
Very long
Few

Small in-process

Large lot

Small lot

Not applicable

Production time
per unit
Examples in
goods industry

Long, variable

Short; constant

Long; variable

Short; constant

Machine shop; tool

and die shop

TV factory; auto
assembly line

Shipbuilding;
house construction

Oil or chemical
processing

Examples in
service industry

Hospital;
restaurant

College
registration;
cafeteria

Movie; TV show;
play; buffet

Agriculture;

80

Black, J. T. (1988) Cellular Manufacturing Systems, in Voss, C. (Editor) (1988) Just-In-Time Manufacture,
IFS (Publications) Ltd UK, Springer-Verlag, pp27-49.

Characteristics of basic manufacturing systems

Automated
Robotic
Assembly Lines

81

JIT and the Manufacturing Environment

Variety
High

JOBBING/PROJECT

Part of JIT
unsuitable, application
of selected techniques BATCH MANUFACTURE
JIT suitable for
part of plant
JIT suitable for
all of plant
LINE MANUFACTURE
JIT suitable for
manufacture PROCESS MANUFACTURE

Medium

Low
Low

Medium

Much of JIT in place,

application of selected
techniques
High

Volume
82
Bicheno, J. R., (1988) A Framework for JIT Implementation, in
Voss, C. (Editor) (1988) Just-In-Time Manufacture, IFS (Publications) Ltd UK, Springer-Verlag, pp 191-204.

Small batch

Lean
manufacturing

Mass
customization

PRODUCT FLEXIBILITY

New choices

Mass
production

Continuous
process

BATCH SIZE

Graph from Richard L. Daft, Jonathan Murphy, Hugh Willmott, Organization Theory, page 280

Cellular Manufacturing
Group dissimilar machines in
manufacturing cell to produce family of
parts
Work flows in one direction through cell (as
much as possible)
One worker tends several machines
Cycle time adjusted by changing number of
workers
84

Requirements of Work Cells

1. Identification of families of products
2. A high level of training and
flexibility on the part of employees
3. Either staff support or flexible,
imaginative employees to establish
work cells initially
4. Test (poka-yoke) at each station in
the cell
85

Source: Heizer & Render

Staffing and Balancing Work Cells

Determine the takt time
total work time available
TAKT time =
units required

Determine the number

of operators required
total operation time
required TAKT time

Workers
required =
86

Source: Heizer & Render

Staffing Work Cells Example

600 Mirrors per day required
Mirror production scheduled for 8 hours per day
From a work balance chart
total operation time = 140 seconds
Standard time required

60

50
40
30

20
10
0 Assemble Paint

Test

Label Pack for

shipment

Operations
87

Source: Heizer & Render

Staffing Work Cells Example

600 Mirrors per day required
Mirror production scheduled for 8 hours per day
From a work balance chart
total operation time = 140 seconds

TAKT time

= (8 hrs x 60 mins) / 600 units

= 0.8 mins = 48 seconds

total operation time required

Workers required =
TAKT time
= 140 / 48 = 2.91

88

Source: Heizer & Render

Manufacturing Cell With Worker

Routes
Machines
Enter

Worker 2
Worker
3

Worker 1

Exit
Key:

89

Product route
Worker route

TAKT Time Observations

Blindly following TAKT calculations will lead to one
solution. This may be only the obvious solution.
Benchmark what competitors do.
Benchmark against an absolute value the absolute
minimum or maximum conceivable.
Volume may increase or drop. Employ flexible workforces
able to take holidays and work when needed.
Work backwards from number of employees available
during holidays.
Increase the process time flexibility reduce time by
identifying efficiencies and reducing component part
count reduces hand time required to pick up and
assemble.
Can the product be made slower with fewer resources?
Outsource low value add work.
90

A Simulation of Lean
Manufacturing

91

Batch Processing Example

Product requires three processes that take one minute each

92

Batch Processing Example

Product requires three processes that take one minute each

93

Batch Processing Example

Product requires three processes that take one minute each

Complete processing of first batch of 10 takes 30 minutes

First good part out takes 21 minutes (plus transport time)
There are at least 21 parts in-process
94

Continuous Flow Processing

Example
Product requires three processes that take one minute each

First part out in 3 minutes

10 completed in 12 minutes
Only 2 sub-assemblies in
process at a time.

95

How MRP, Kanban and Heijunka

Fit Together

Source: D.Ralston, IOM Control

(http://www.littoralis.info/iom/secure/assets/19960601b.pdf)
96

Forrester and Burbidge Effects

Amplified demand and low synchronisation

Factory orders
Forrester Effect
Actual demand
variation

Mean

Actual Demand
Fluctuation

Effect of poor information flow and multi-phase ordering in a supply chain

Factory orders
Burbidge Effect

Mean

Forrester, J.W. (1961), Industrial Dynamics, MIT Press.

Burbidge, J. (1995) Back to production management,
97Manufacturing Engineer, 74 (2) pp66-71.
Naim, M. (1997) The book that changed the world, Manufacturing Engineer, IEE, February 1997, pp13-16.

The Consultants Flywheel Explanation of

Supply Chain Demand Amplification

Apparent
Shortages

Real

Demand
distortion

Safety
stock
increase

98

Time
Over
ordering

Unreliable
delivery

Causes of uneven demand at plant level

DC1

DC2

DC3

DC4

Independent 1 1 1 1 1 1 1
Demand

5 5 5 5 5 5 5

3 3 3 3 3 3 3

11 11 11 11 11 11 11

Orders
Placed (A)

15

10

15

15

10

10 25

RDC 1

25

25

25

RDC 2

Combined
Demand

20

15

5 15

35

25 10 25

35

Orders
Placed (B)

25

25

25

50

50

50

PLANT

Orders
Received

75

50 25

75

(A) Lot size replenishment policies in multiples of 5 (overpack quantity)

(B) Lot size replenishment policies in multiples of 25 (pallet/truck)

Economic order quantities and economic batch quantities

99

Christopher, M. (1997)

Idealised Make to Replenish

DC1
Independent 1 1 1 1 1 1 1
Demand

Orders
Placed (A)

DC2

DC3

DC4

5 5 5 5 5 5 5

3 3 3 3 3 3 3

11 11 11 11 11 11 11

1 1 1 1 1 1 1

5 5 5 5 5 5 5

3 3 3 3 3 3 3

RDC 1

11 11 11 11 11 11 11

RDC 2

Combined
Demand

6 6 6 6 6 6 6

14 14 14 14 14 14 14

Orders
Placed (B)

6 6 6 6 6 6 6

14 14 14 14 14 14 14
PLANT

Orders
Received

20 20 20 20 20 20 20

(A) Lot size replenishment policies - equals daily sales volume

(B) Lot size replenishment policies equals regional consolidated actual replenishment volume

100

Forrester Effect

Period
1
2
3
4
5
6

Third Tier
Second Tier
First Tier
Original
Supplier
Supplier
Supplier
equipment mfg
Prodn Stock Prodn Stock Prodn Stock Prodn Stock Demand
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
-60
20
20
60
60
80
80
90
90
20
60
80
90
260
140
140
100
100
90
90
90
90
140
100
90
90
20
80
80
90
90
90
90
90
90
80
90
90
90
100
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
101

Forrester Effect

300
250

Fluctuations

200
150
100
50
0
1

-50

Period

-100
Demand

OEM Prd
102

1st Tier

2nd Tier

3rd Tier

Forrester Effect
Period
1
2
3
4
5
6

Supplier
Supplier
Supplier
equipment mfg
Prodn Stock Prodn Stock Prodn Stock Prodn Stock Demand
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
250
170
170
130
130
110
110
100
100
170
130
110
100
-70
50
50
90
90
100
100
100
100
50
90
100
100
170
110
110
100
100
100
100
100
100
110
100
100
100
90
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100

Source: After Slack et al103

Damped fluctuations of production levels along a supply

chain in response to small change in end-customer
Third Tier
Second Tierdemand
First Tier
Original
Period
1
2
3
4
5
6

Supplier
Supplier
Supplier
equipment mfg
Prodn Stock Prodn Stock Prodn Stock Prodn Stock Demand
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
90
90
100
100
100
90
100
100
100
100
100
90
90
90
90
100
100
90
90
100
100
100
90
90
90
90
90
90
100
90
90
90
100
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90

The only change between these two tables is from Original stock minus end stock to End
stock minus original stock plus production. This allows for a buffer to be created if demand
increases. Discuss the merits of
104the logic in both tables.

Damped fluctuations of production levels along a supply

chain in response to small change in end-customer
Third Tier
Second Tierdemand
First Tier
Original
Period
1
2
3
4
5
6

Supplier
Supplier
Supplier
equipment mfg
Prodn Stock Prodn Stock Prodn Stock Prodn Stock Demand
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
100
100
90
90
90
100
90
90
90
90
90
100
100
100
100
90
90
100
100
90
90
90
100
100
100
100
100
100
90
100
100
100
90
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100

105

Forrester Effect permanent reduction in demand

Third Tier
Second Tier
First Tier
Original
Supplier
Supplier
Supplier
equipment mfg
Period Prodn Stock Prodn Stock Prodn Stock Prodn Stock Demand
100
100
100
100
1
100
100
100
100
100
100
100
100
100
100
100
100
100
2 50 -60
20 60 20
60 70 60
80
80
90
90
90 20
90 60
90 80
90
3 90 260
140 90 140
100 90 100
90
90
90
90
90
90
90 140
90 100
4 90 20
80 90 80
90
90
90
90
90
90
90
90
90
90 80
5 90 100
90
90
90
90
90
90
90
90
90
90
90
90
6
90
90
90
90
90
90
90
90
90
106

Source: Analysis by F. Goethals