Scribd
Upload
25 views15 pages

Optimization Lec04 LinearProgramming

The document discusses linear programming problems and their solutions. It introduces linear programming, describing how linear problems can be solved and how non-linear problems are converted to linear ones. It then covers problem formulation, showing the standard form of a linear programming problem. Finally, it discusses the graphical method for solving 2D linear programming problems visually by finding the optimal corner point of the feasible region.

Uploaded by

John Lamei
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
25 views15 pages

Optimization Lec04 LinearProgramming

The document discusses linear programming problems and their solutions. It introduces linear programming, describing how linear problems can be solved and how non-linear problems are converted to linear ones. It then covers problem formulation, showing the standard form of a linear programming problem. Finally, it discusses the graphical method for solving 2D linear programming problems visually by finding the optimal corner point of the feasible region.

Uploaded by

John Lamei
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 15

02-24-01410 - Non-Linear

and Combinatorial
Optimization
DR. AHMED TAYEL
De p a rtment o f E n gi neering M at h em ati cs a n d P hysi cs, Fa c ul ty o f
E n gi neering , Al exandria Un i ve rsi ty
a h m e d .taye l @ a l ex u . e d u . e g
Linear Programming
1. Introduction.
2. Graphical method.
3. Simplex method.
1. Introduction
Why linear programming?

• Very easy to solve.

• There exist efficient algorithms to solve large linear programming problems.

• Non-linear problems can be solved by conversion to an equivalent linear

programming problem (Frank-wolf algorithm).


Problem formulation

Max 𝑧 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + ⋯ + 𝑐𝑛 𝑥𝑛

Subject to 𝑎11 𝑥1 + 𝑎12 𝑥2 + ⋯ + 𝑎1𝑛 𝑥𝑛 ≤ 𝑏1


𝑎11 𝑥1 + 𝑎12 𝑥2 + ⋯ + 𝑎1𝑛 𝑥𝑛 ≤ 𝑏1 Functional
⋮ constraints

𝑎11 𝑥1 + 𝑎12 𝑥2 + ⋯ + 𝑎1𝑛 𝑥𝑛 ≤ 𝑏1


Non-negativity
𝑥1 , 𝑥2 , … , 𝑥𝑛 ≥ 0
constraints

Handling variables unrestricted in sign


𝑥𝑖 = 𝑥𝑖+ − 𝑥𝑖−
𝑥𝑖+ , 𝑥𝑖− ≥ 0
Intuition about the solution

Definition:
Corner point solution:
Point resulting from the
intersection of two or more
constraint boundaries.
Theorem:
The optimal solution(s) lie
on one of the corner points
of the feasible region.
Solution to the linear
programming problem

Graphical solution Algebraic solution


(Simplex Algorithm)
• Accuracy issue.
• Handles only 2D problems. • Very accurate.
• Handles 𝑛 −D problems (𝑛 ≥ 2).

This lecture
Next lecture
2. Graphical solution
𝑦
Example: Max 𝑧 =3𝑥+2𝑦
Subject to 2𝑥+3𝑦≤6
2𝑥+𝑦 ≤4
𝑥≤4
𝑥, 𝑦 ≥ 0

Point 𝒛
(0,0) 0 𝑥
(2,0) 6
(1.5,1) 6.5
(0,2) 4
𝑦
Example: Max 𝑧 =2𝑥+ 𝑦
Subject to 𝑥+𝑦≤3
2 𝑥 + 3 𝑦 ≥ 12
𝑥, 𝑦 ≥ 0

𝑥
𝑦
Example: Max 𝑧 =2𝑥+3𝑦
Subject to 2𝑥+3𝑦≤6
𝑥+2𝑦 ≤6
𝑥, 𝑦 ≥ 0

Point 𝒛
(0,0) 0 𝑥
(3,0) 6
(0,2) 6
Example: Max 𝑧 =3𝑥+2𝑦 𝑦
Subject to 𝑥 − 2 𝑦 ≤ 10
2 𝑥 − 𝑦 ≤ 10 Unbounded
𝑥, 𝑦 ≥ 0 feasible region

𝑥
𝑦
Example: Min 𝑧 =2𝑥+3𝑦
Subject to 2𝑥+𝑦 ≥7
𝑥+𝑦≥6 Unbounded
feasible region
𝑥+4𝑦 ≥8
𝑥, 𝑦 ≥ 0

Point 𝒛
(0,7) 21 𝑥
(1,5) 17
(16/3,2/3) 38/3 = 12.67
(8,0) 16

You might also like