0% found this document useful (0 votes)
7 views22 pages

OR 12thweek

The document discusses network flow models in operations research, highlighting their importance in managing complex supply chains, such as Proctor & Gamble's extensive network. It outlines the general formulation of these models, including concepts like directed networks, flows, capacities, and various types of problems such as minimum cost flows and maximum flow. Additionally, it references key literature in the field to support the concepts presented.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
7 views22 pages

OR 12thweek

The document discusses network flow models in operations research, highlighting their importance in managing complex supply chains, such as Proctor & Gamble's extensive network. It outlines the general formulation of these models, including concepts like directed networks, flows, capacities, and various types of problems such as minimum cost flows and maximum flow. Additionally, it references key literature in the field to support the concepts presented.
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

NETWORK FLOW MODELS

ISL323E- OPERATIONS RESEARCH


DR. AZİZ KEMAL KONYALIOĞLU
15th Dec, 2025
konyalioglua@[Link]
Class Hours: Monday 14.30/17.30
Tuesday 14.30/17.30
CONTENT FOR TODAY

¡ Why is it important?
¡ General formulation of network-flow models
¡ Examples
NETWORK FLOW MODELS

¡ Proctor & Gamble makes and markets over 300 brands of consumer goods worldwide.
¡ In the past, P & G had hundreds of suppliers, over 60 plants, 15 distributing centers, and over 1000 consumer
zones.
¡ Managing item flows over the huge supply network is challenging!
¡ An LP/IP model helps.
¡ The special structure of network transportation must also be utilized.
¡ $200 million are saved after an OR study!
NETWORK FLOW MODELS

¡ A lot of operations are to transport items on a network.


¡ Moving materials from suppliers to factories.
¡ Moving goods from factories to distributing centers.
¡ Moving goods from distributing centers to retail stores.
¡ Sending passengers through railroads or by flights.
¡ Sending data packets on the Internet.
¡ Sending water through pipelines and many more.
¡ A unified model, the minimum cost network flow (MCNF) model, covers many network operations.
¡ It has some very nice theoretical properties.
¡ It can also be used for making decisions regarding inventory, project management, job assignment, facility location, etc.
NETWORK FLOW MODELS

¡ A network (graph) has nodes (vertices) and arcs (edges/links).


¡ A typical interpretation: Nodes are locations and arcs are roads.
¡ Arcs may be directed or undirected.
¡ For an arc from u to v: (u, v) if directed and [u, v] if undirected.
¡ In this lecture, all arcs are directed.
¡ A network is directed if its arcs are directed.
¡ An undirected network is also called a graph (by some people).
¡ A path (route) from node s to node t is a set of arcs
¡ (s, v1), (v1, v2), ..., (vk−1, vk), and (vk, t) such that s and t are connected.
¡ ) A cycle (equivalent to circuit in some textbooks) is a path whose destination node is the source node.
¡ ) A path is a simple path if it is not a cycle.
¡ ) A network is an acyclic network if it contains no cycle.
NETWORK FLOW MODELS
Flows, weights, capacities

¡ A flow on an arc is the action of sending some items through the arc.
¡ The number of units sent is called the flow size.
¡ A network flow is the collection of all arc flows.
¡ A network flow is just a plan for making flows on all arcs.
¡ An arc may have a weight.
¡ A weight may be a distance, a cost per unit flow, etc.
¡ A weighted network is a network whose arcs are weighted.
¡ An arc may have a capacity constraint.
¡ There may be an upper bound and/or an lower bound (typically 0) for its flow size.
¡ A network is capacitated if there is an arc having capacity limits.
NETWORK FLOW MODELS
NETWORK FLOW MODELS

m!n mol!gette g!t


demand
object!ve func
.
eepes!!md!yet
te - Xe 2
≤ 12

θ X13[20
4Xe2 + 3xez + ----o 亠 X
λ 23

24 ≤ 12
≤ 2o

s t
X
-

^ 25

ç!za-g!ren
\
×
z4
M
X
X
12 + X13
- 0 = 25 demand
35

X 45
_

10 X53
445
-

*
y34
=
+ -

84

_
MINIMUM COST NETWORK FLOWS

+ 1
-
9
=
ottrassh!pment danad el!m!zde yönlü , kapas!tel! ve ağırlıklı b!r ağ var .

G =
(0 , E)
25> o
t
=d!s!ns
supply
denand E =
yaylar ,
kanalar

Duğum türler!

b! s o
tsupply node

b! < O- demand node


25 =
( ot (
b!= o - transh!pmentnode
Y
problem freas!ble
Ʃ b! = o

yaylar :
(v!j ,
< !j) ! EO
dd
Kopos!te b!r!m mal!yet
l!st sur) toplan arzz toplan talep

örnez : (1 + 2) 15
:

, $) '

)
r -
<$
>(: 20 ,
3 )
2 + 3) : 10 e


dec!s!on var!ables
X!j = (! >-

j) yayı üzer!ndek! akış m!ktarı

tüm (!jj) & E !ç!n tanımlanır .

amaç fonks!yonu

4X12 -X 13 + 2x 23 +2x +3X


m!n
24 25
+
2x34 +14
25
+
2x45 4X38 +

5 katsayılar b!r!m mal!yetler .


model!n tom matemat!ksel formu

m! Gx 12 + Bx 1y + 2x2y + 2x24 + 2x25 + 2x34 + X35 + 2x45 + Gx5]


s!t
Kapos!te surler

X12 + X13 = 25

o X!j[V!j & X12[15 -

X12 +
x23 +
x24
+
X25 = 0

×5 0
× t
434 t × } =
X13[20
-

13 423 35
- -

X 5315
×
2みー × 3み ← ×
み5
ニー (O

35 *
-

+ x =
17
-

Flow Bolance Constra!nts 45 5z

kurd Çıkan akış akış b! O X!jV!j (! , EE


gene g!ren
-

d!sum (transsh!pment bod dujum 5 (demand b5 15)


d!sum 1/ supply , b , 25) 3 , = -

= ,
·

*5 + X 53 15
*
85435
-
-
= -

+ X
125 × 13
*
13 x + X
x5z = 0
-

23 34 35
-

X
-

= 25

duğum G (demand -10)


·
d!sum 2 ( transsh!pment , be = o) , bu =

x24
-

-
x34 +
Xe = -
10
-

X12 + X23 + X24 +


y25 =
0

MINIMUM COST NETWORK FLOWS
'
MINIMUM COST NETWORK FLOWS
MINIMUM COST NETWORK FLOWS
ASSIGNMENT PROBLEMS
TRANSSHIPMENT PROBLEMS
SHORTEST PATH PROBLEMS
MAXIMUM FLOW PROBLEMS
AN OVERVIEW
AN EXAMPLE (SHORTEST ROUTE)
!zer!nde
sen!n
X!j
31 efe phetestpoth
=

1
m!nz =
Ec!px!s
d
_
Cost- '
g!o! \
düşün
1
m!nz = 100 x12 + 30x13 + 20 x23 + 1034 +
15x42 + 68x35 + 5045 ←

· r!r !
source

exte trassn!pment g!rel = G!zen

nodes +
X12 +
113 = 1 b!r! seç!len
- 2
yolda
&
Rode2 +
X12 + X12 =
X2 + 1 + hade oldağo
!ç!p
1
!nputs g!ren-alken =

mode3 - X13 + X23 = X34 + X35 >


- hedel değ"l -
g!ren = alde

mode 4 Xu2
->
Xzu = + X
+5

nodes =
Xg5 + Xu5 =
0 g!ren = sl!e - O
L!ndgee
AN EXAMPLE (TRANSSHIPMENT)
supp □
θ
%* We denends
>
-
Horp
ง, I

tümem
% 0
β
0000000000
melt
J el!m!zbekenstme!nt !

y frels
akan
- zeraa
g!ren
>
- zerana g!ren çıka

] g!ver
replen deverd!
-

kers " enel


"
จั๋
MAXIMUM FLOW

!ntere

=
ab!z
& at!f!c!al
o
top!e ,
dovete
s!ndle g!ve
dd
sourcecenter
Xo
=

QO nebel
b!z!m Xo

ara nodeler 1-2-3



nodef
X80 , X 1
S = X
2+ ,3

m!ktarı g!ren alzan


S03 borudan
geçen petrol
source

S!- s!nz
X!j >

rode 2

]
XS0 , 2 ≤2
Kop ps!te
en fazla kon !ş * + X
c, 2 s012 =
X25!
göndereb!l!r!m
1

y ,
34 sınırsanı node3
X3 s! [1 X2 3
,

X3 , s!
1

=
,

Xo = X50 , 1 + Xso , 2 toplan G!ze

X3 ,
s! +
X2 , s! = Xo - toplan g!ren
REFERENCES

¡ Taha, H. A. (2013). Operations research: an introduction. Pearson Education India.


¡ Hillier, F. S., & Lieberman, G. J. (2015). Introduction to operations research. McGraw-Hill,.
¡ Bazaraa, M. S., Jarvis, J. J., & Sherali, H. D. (2011). Linear programming and network flows. John Wiley & Sons.
¡ Massachusetts Institute of Technology, Mathematics Department, Operations Research Class Notes
¡ DELFT University of Technology, Mathematics Department, Operations Research Class Notes
¡ Taiwan Technical University, Ling-Chieh Kung’s Class Notes

You might also like