Scribd
Upload
137 views

Models For The Layout Problem

The document discusses various models for layout problems, including single row, multi-row, and loop layout problems. It introduces the ABSMODEL 1, ABSMODEL 2, and ABSMODEL 3 models for single row, multi-row with equal sized facilities, and multi-row with unequal sized facilities respectively. It also discusses the quadratic assignment problem (QAP) model for multi-row layouts and introduces the linear mixed integer programming (LMIP-2) model for modeling loop layout problems. Examples are provided to illustrate how these models can be formulated and solved using software.

Uploaded by

Akamai Corporation
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
137 views

Models For The Layout Problem

The document discusses various models for layout problems, including single row, multi-row, and loop layout problems. It introduces the ABSMODEL 1, ABSMODEL 2, and ABSMODEL 3 models for single row, multi-row with equal sized facilities, and multi-row with unequal sized facilities respectively. It also discusses the quadratic assignment problem (QAP) model for multi-row layouts and introduces the linear mixed integer programming (LMIP-2) model for modeling loop layout problems. Examples are provided to illustrate how these models can be formulated and solved using software.

Uploaded by

Akamai Corporation
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 61

Models for The Layout

Problem
Layout Problem
 The objective is to minimize the cost or time in:
 Man and material’s movement between offices
 Transporting work in process, finished parts, materials, and tools
between machines or workstations
 Picking parts from storage spaces

 Classify the layout problem into 2 types :


 Single row layout problem
 One dimensional
 ABSMODEL 1
 Multirow layout problem
 Two dimensional
 ABSMODEL 2, QAP
Single Row Layout
Multirow Layout
Model for the Single Row Layout
Problem
 Assumptions :
 The facilities are square or rectangular
 The facilities are arrange along a straight line
 The orientation of facilities is known a priori
 There is no restriction on the shape of building in
which the facilities are to be housed
Approximation of Facility shapes that
are not exact squares or rectangles
Approximation of Facility shapes that
are not exact squares or rectangles
Treating L Shapes Facility as Rectangular
Treating L Shapes Facility as Square
ABSMODEL 1
 n = Number of facilities in the problem
 Cij = Cost of moving standard unit by a unit distance
between facilities i and j
 f ij = Number of trips between facilities i and j

 li = Length of the horizontal side of facility i


d
 ij = Minimum distance by which facilities i and j
are to be separated horizontally
 H = Horizontal dimension of the floor plan
ABSMODEL 1
 xi = distance between the center of facility i and the
Vertical Reference Line (VRL)
 minimize
n 1 n

 C
i 1 j i 1
ij f ij | xi  x j |

 subject to
1
| xi  x j |  (li  l j )  d ij , i  1, 2, ..., n  1, j  i  1, ..., n
2
Illustration of Parameters and Decision Variables
for the Single Row Layout Problem
d ij
li lj

Facility j
Facility i

xi
xj

VRL
ABSMODEL 1
 If the building dimension is known and the
user wants to include the constraint that the
facilities are located within the horizontal
dimension, the constraint should be added :

1 1
H  li  xi  li i  1, 2, ..., n
2 2
xj

xi

Facility i Facility j
...

H
li lj
VRL
Customer
Service
General Repair Part Display
Area Area
Room
1 2 3 4 5

1 - 12 8 20 0
Room

[ f ij ]  2 12 - 4 6 2

3 8 4 - 10 0

4 20 6 10 - 3

5 0 2 0 3 -
Room Dimensions
Room Name
Number (in feet)

1 TV/VCR 20 x 10

2 Audio 10 x 10

3 Microwave 10 x 10

4 Computer 20 x 10

5 Parts 15 x 10
Analysis of ABSMODEL 1 Using GINO
FINAL LAYOUT:[TVCR]
MODEL FOR THE MULTI-ROW
LAYOUT PROBLEM
 The control panel layout problem; The machine
layout problem; Typewriter keyboard design; Office
layout (Burkard, 1984).
 Various Models have been developed to formulate
the problem.
 QAP (Quadratic Assignment Problem) –(Koopman
and Beckman, 1957)
 Linear mixed integer programming (Love and Wong,
1976)
 The quadratic set covering model (Bazaraa, 1975)
 The non linear model (Heragu, 1992)
QAP
(Quadratic Assignment Problem)
 n facilities to be assigned to m locations
 Assignment means matching each facility
with a specific location and vice versa.
 If m<n, use dummy (n-m) locations and
assign zero flow
 If n<m, the problem is infeasible
 QAP=Assignment Problem in which the
objective function is quadratic
Four sites, Four facilities and the
assignment of the facilities to the sites

1 2 a b b,1 d,2

3 4 c d c,3 a,4
Mathematical Model:[QAP]
Mathematical Model:[QAP]
Mathematical Model:[QAP]
QAP
 Optimal algorithms: Branch and bound,
Decomposition, Cutting plane algorithms
 These algorithms can produce optimal
solutions for only small sized problem
(M<=20)
 Since it is computationally infeasible to find
an optimal solution when M>20.
 Heuristic (non-optimal) procedures are used
to solve the problem – e.g. Steepest Descent
Pairwise Interchange (SDPI) Heuristic
EXAMPLE 2:LonBank Inc
 Four square offices of equal area are to be
located in the new square building
 Formulate the problem as a QAP using data in
Fig. 5.15
 The site in the building are shown in Fig. 5.16
 This problem is solved using GINO.
Inputs required by GINO
SOLUTION: LonBank Inc
ABSMODEL 2
 ABSMODEL 1
 The single-row facility layout problem
 ABSMODEL 2
 Two or more rows
 Not even along straight rows
 Facilities are squares of equal area
ABSMODEL 2
 Assumptions:
 No clearance between the facilities
 The facilities have a unit dimension –length and
width are 1 unit each
 Constraints:
 (11) and (12) ensure that no facilities in the layout
overlap
ABSMODEL 2
Decision Variables and Reference
Lines –HRL & VRL
Decision Variables and Reference
Lines –HRL & VRL
SOLUTION:[Using GINO]
MODEL FOR THE MULTI-ROW LAYOUT PROBLEM WITH
FACILITIES OF UNEQUAL AREA

 The model is called ABSMODEL 3


 The assumptions –equal area and square-
shape are relaxed
 Assumptions:
 Facilities are square or rectangular
 Physical orientation is known a priori
ABSMODEL 3
Parameters and Decision variables
Mathematical
Model:[ABSMODEL 3]
EXAMPLE 4: [Insurance Co.]
 To determine the location of five office
cubicles of varying dimensions
 The estimated trips and office dimensions
(See Fig. 5.24)
 Use ABSMODEL 3 to formulate the problem
LOOP LAYOUT PROBLEM
 To formulate a layout problem that arises
commonly in automated manufacturing – the
loop layout problem (LLP)
 The LLP determines the optimal arrangement
of machines or workstations around a loop
(See Fig. 5.27)
LOOP LAYOUT
LOOP LAYOUT

Part Family W Part Family X

Part Family Y Part Family Z


Cellular Layout HM

VM

1. Identify families of Worker 3


VM
parts with similar
flow paths L

2. Group machines
into cells based on
part families Worker 2
G
3. Arrange cells so
L
material movement
is minimized Final
inspection

4. Locate large shared


machines at point of Finished
part
S Worker 1
use
In Out
JIT LAYOUT
Lecture-05
LOOP LAYOUT
 A typical loop layout has a pick up and drop-
off point through which all parts enter and
exit the system
 Only the internal loop is used for material
handling –loop conveyor, tow line, overhead
monorail system, or wire paths of AGVs
 The movement of materials may be
unidirectional or bidirectional
 Unidirectional loop layout are preferred (in
Japan)
LMIP-2
 LMIP (Linear Mixed Integer Programming)
 LMIP 2:
 Not assume that candidate sites are known a
priori
 Suitable for unidirectional flow, bidirectional
flow, backtracking, and no backtracking.
LLP
 We assume that the LLP is equivalent to the
single-row layout problem
 If the material movement is unidirectional,
then the LLP is equivalent to the single-row
layout problem with no backtracking
 In model LMIP 2, we use the following
notations and assume that material
movements occur in a prespecified direction
(i.e. unidirectional)
Mathematical Model:[LLP]

You might also like