indian institute of science - nptelnptel.ac.in/courses/105108130/module2/lecture09.pdf · ·...
TRANSCRIPT
![Page 1: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/1.jpg)
Water Resources Systems:Modeling Techniques and Analysis
Lecture - 9Course Instructor : Prof. P. P. MUJUMDAR
Department of Civil Engg., IISc.
INDIAN INSTITUTE OF SCIENCE
1
![Page 2: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/2.jpg)
Summary of the previous lecture
2
• Graphical method• General form of LP• Motivation for the simplex method
![Page 3: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/3.jpg)
LP – Simplex Method
Motivation for the Simplex method
Max Z =
s.t.
3
1 26 8x x
1 25 10 60x x
1 2
1
2
4 4 40
0
0
x x
x
x
![Page 4: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/4.jpg)
4
x1
x2
LP – Simplex Method
Feasible region
O D
A
B
C
E
![Page 5: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/5.jpg)
LP – Simplex Method
In the general form of LP, the constraints are converted as
5
1 25 10 60x x
1 2
1
2
4 4 40
0
0
x x
x
x
1 2 35 10 60x x x
1 2 4
1 2
3 4
4 4 40
0; 0
0; 0
x x x
x x
x x
2 equations and 4 unknowns
x3, x4 are slack variables
![Page 6: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/6.jpg)
LP – Simplex Method• In general, m equations in n unknowns (n > m),
including slack and surplus variables• It is possible to solve for m variables in terms of the
other (n – m) variables• Basic solution: A basic solution is a solution obtained
by setting (n – m) variables to zero• The m variables whose solution is sought by setting
the remaining (n – m) variables to zero are called the basic variables
• The (n – m) variables which are set to zero are called the non-basic variables. It may so happen that some variables may take a value of zero arising out of the solution of the m equations, but they are not non-basic variables as they are not initialized to zero.
6
![Page 7: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/7.jpg)
LP – Simplex Method
• Since out of n variables, any of the (n – m) variables may be set to zero, the no. of basic solutions will correspond to the no. of ways in which m variables can be selected out of n variables i.e.,
For the problem being considered, with n = 4 and m = 2, the no. of basic solutions will be
7
mnC
!! !
nm n m
4! 6
2! 4 2 !
=
![Page 8: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/8.jpg)
8
x1
x2
LP – Simplex Method
Feasible region
O D
A
B
C
E
![Page 9: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/9.jpg)
LP – Simplex Method
• In the example, the basic solutions are
9
x1 x2 x3 x4 Point on graph Feasible0 0 60 40 O Y0 6 0 16 A Y0 10 -40 0 B N8 2 0 0 C Y10 0 10 0 D Y12 0 0 -8 E N
1 2 35 10 60x x x 1 2 44 4 40x x x
![Page 10: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/10.jpg)
10
LP – Simplex Method
• All the basic solutions need not (and, in general, will not) be feasible.
• A basic solution which is also feasible is called as the Basic Feasible Solution.
• All the corner points of the feasible space are basic feasible solutions.
![Page 11: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/11.jpg)
11
LP – Simplex Method
• As the size of a LP problem increases, the no. of basic solutions also increases.
• The possible no. of basic feasible solutions can be too large to be enumerated completely.
• The goal would be, starting with an initial basic feasible solution, to generate better and better basic feasible solutions until the optimal basic feasible solution is obtained.
![Page 12: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/12.jpg)
12
LP – Simplex MethodAlgebraic approach (Simplex algorithm):
• To solve the problem algebraically, we must be able to sequentially generate a set of basic feasible solutions
1 2 35 10 60x x x
1 2 4
1 2
3 4
4 4 40
0; 0
0; 0
x x x
x x
x x
1 2 3 46 8 0 0Z x x x x Maximize
s.t.
![Page 13: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/13.jpg)
13
LP – Simplex Method
1. Determine an initial basic feasible solution:• The best way to obtain an initial basic feasible
solution would be to put all the decision variables (n – m in no., if we have one slack/surplus variable associated with each constraint) to zero.
• In the example, choose the slack variables x3 and x4to be basic and x1 and x2 to be non-basic (i.e., set x1 = 0 and x2 = 0 )
Basis : 3
4
xx
1 2 35 10 60x x x
1 2 44 4 40x x x
![Page 14: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/14.jpg)
14
LP – Simplex Method
• Solve the m (=2) equations for m basic variables
1 2 35 10 60x x x
1 2 44 4 40x x x 3 60x
4 40x
Initial basic feasible solution is3
4
6040
xx
Point ‘O’ on the graph
Z = 0
![Page 15: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/15.jpg)
15
LP – Simplex Method
• Now we have to generate another basic feasible solution from the initial basic feasible solution such that there is an improvement in the objective function value, Z.
• The new solution is obtained as follows• ONE presently non-basic variable (x1 or x2) must
be selected to enter the current basis, thus becoming basic.
• ONE presently basic variable (x3 or x4) must be selected to leave the current basis, thus becoming non-basic.
![Page 16: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/16.jpg)
16
LP – Simplex Method
First operation:
• Either x1 or x2 can enter the basis.• The question now is, out of x1 and x2 which variable
should enter the basis?• Since the problem is to maximize the objective
function (OF), we must look for the variable which will increase the OF value the fastest.
• Because the coefficient of x2 in the OF is higher than that of x1, x2 increases the OF value faster than x1does. Hence x2 should enter the basis.
1 2 3 46 8 0 0Z x x x x
Current basis :3
4
xx
![Page 17: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/17.jpg)
17
LP – Simplex Method
Second operation:• As the value of newly selected basic variable x2 is
increased (x1 still being zero), the other two variables, x3 and x4 (which are currently in the basis) keep reducing.
• One of these will reach its lower limit (zero) earlier than the other
• x3 = 0 when 10x2 = 60 or x2 = 6• x4 = 0 when 4x2 = 40 or x2 = 10
3 260 10x x
4 240 4x x
1 2 35 10 60x x x
1 2 44 4 40x x x
![Page 18: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/18.jpg)
18
LP – Simplex Method• To increase the OF value, we would be interested in
increasing x2 to the greatest extent possible.• As soon as x2 reaches 6, x3 becomes zero (x4 is still
non-zero)• Therefore for x2 entering the basis, x3 must leave the
basis.• Therefore new basis is
• The new basic solution is obtained by Guass Jordan elimination method.
• The process continues until the maximum value of the function is arrived at.
4
2
xx
![Page 19: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/19.jpg)
Example – 1Maximize
s.t.
19
Constraints
Decision variables
21 53 xxZ
00
1823122
4
2
1
21
2
1
xx
xxx
x
![Page 20: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/20.jpg)
0;0;00;0
1823122
4
543
21
521
42
31
xxxxx
xxxxx
xx
Example – 1 (Contd.)The problem is converted into standard LP form
s.t.
20
n = no. of variables = 5; m = no. of constraints =3
1 2 3 4 53 5 0 0 0Z x x x x x Maximize
00
1823122
4
2
1
21
2
1
xx
xxx
x
![Page 21: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/21.jpg)
Example – 1 (Contd.)
Seek solution from m variables by setting (n – m) variables to zero
The solution is obtained by creating a table with coefficients of variables.
21
x1 and x2 are non basic variables (whose values are set to zero)
x3, x4 and x5 are basic variables (whose solution is sought)
![Page 22: INDIAN INSTITUTE OF SCIENCE - NPTELnptel.ac.in/courses/105108130/module2/Lecture09.pdf · · 2017-08-04INDIAN INSTITUTE OF SCIENCE 1. Summary of the previous lecture 2 ... • The](https://reader036.vdocuments.mx/reader036/viewer/2022062600/5ac3ac7b7f8b9ae06c8c6674/html5/thumbnails/22.jpg)
Example – 1 (Contd.)Iteration-1
22
Basis Row Z x1 x2 x3 x4 x5 bi
Z 0 1 -3 -5 0 0 0 0
x3 1 0 1 0 1 0 0 4
x4 2 0 0 2 0 1 0 12
x5 3 0 3 2 0 0 1 18
1 2 3 4 53 5 0 0 0 0Z x x x x x
1 3
2 4
1 2 5
42 123 2 18
x xx xx x x