linear programming - university of washingtonburke/crs/407/lectures/l2-introduction.pdf · what is...
TRANSCRIPT
![Page 1: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/1.jpg)
Linear Programming
Lecture 2: Introduction to Linear Programming
Lecture 2: Introduction to Linear Programming Linear Programming 1 / 46
![Page 2: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/2.jpg)
1 Math 407: Introduction
2 What is linear programming?
3 Applications of Linear Programing
4 Example: Plastic Cup Factory
5 Introduction to LP Modeling
6 Graphical Solution of 2D LPs
7 Introduction to Sensitivity Analysis
8 The Theory of Linear Economic ModelsProduction ModelsThe Optimal Value Function and Marginal ValuesDuality: The Hidden Hand of the Market Place
9 LP DualityThe Weak Duality Theorem of Linear Programming
Lecture 2: Introduction to Linear Programming Linear Programming 2 / 46
![Page 3: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/3.jpg)
What is optimization?
A mathematical optimization problem is one in which some function is eithermaximized or minimized relative to a given set of alternatives.
The function to be minimized or maximized is called the objective function.
The set of alternatives is called the feasible region (or constraint region).
In this course, the feasible region is always taken to be a subset of Rn (realn-dimensional space) and the objective function is a function from Rn to R.
Lecture 2: Introduction to Linear Programming Linear Programming 3 / 46
![Page 4: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/4.jpg)
What is optimization?
A mathematical optimization problem is one in which some function is eithermaximized or minimized relative to a given set of alternatives.
The function to be minimized or maximized is called the objective function.
The set of alternatives is called the feasible region (or constraint region).
In this course, the feasible region is always taken to be a subset of Rn (realn-dimensional space) and the objective function is a function from Rn to R.
Lecture 2: Introduction to Linear Programming Linear Programming 3 / 46
![Page 5: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/5.jpg)
What is optimization?
A mathematical optimization problem is one in which some function is eithermaximized or minimized relative to a given set of alternatives.
The function to be minimized or maximized is called the objective function.
The set of alternatives is called the feasible region (or constraint region).
In this course, the feasible region is always taken to be a subset of Rn (realn-dimensional space) and the objective function is a function from Rn to R.
Lecture 2: Introduction to Linear Programming Linear Programming 3 / 46
![Page 6: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/6.jpg)
What is optimization?
A mathematical optimization problem is one in which some function is eithermaximized or minimized relative to a given set of alternatives.
The function to be minimized or maximized is called the objective function.
The set of alternatives is called the feasible region (or constraint region).
In this course, the feasible region is always taken to be a subset of Rn (realn-dimensional space) and the objective function is a function from Rn to R.
Lecture 2: Introduction to Linear Programming Linear Programming 3 / 46
![Page 7: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/7.jpg)
What is optimization?
A mathematical optimization problem is one in which some function is eithermaximized or minimized relative to a given set of alternatives.
The function to be minimized or maximized is called the objective function.
The set of alternatives is called the feasible region (or constraint region).
In this course, the feasible region is always taken to be a subset of Rn (realn-dimensional space) and the objective function is a function from Rn to R.
Lecture 2: Introduction to Linear Programming Linear Programming 3 / 46
![Page 8: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/8.jpg)
What is linear programming (LP)?
A linear program is an optimization problem in finitely many variableshaving a linear objective function and a constraint region determinedby a finite number of linear equality and/or inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 4 / 46
![Page 9: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/9.jpg)
What is linear programming (LP)?
A linear program is an optimization problem in finitely many variableshaving a linear objective function and a constraint region determinedby a finite number of linear equality and/or inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 4 / 46
![Page 10: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/10.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 11: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/11.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 12: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/12.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 13: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/13.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 14: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/14.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 15: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/15.jpg)
What is linear programming (LP)?
A linear program is an optimization problem
in finitely many variables
having a linear objective function
and a constraint region determined by afinite number of constraints
that are linear equality and/or linear inequality constraints.
Lecture 2: Introduction to Linear Programming Linear Programming 5 / 46
![Page 16: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/16.jpg)
A linear function of the variables x1, x2, . . . , xn is any function f of the form
f (x) = c1x1 + c2x2 + · · ·+ cnxn
for fixed ci ∈ R i = 1, . . . , n.
A linear equality constraint is any equation of the form
a1x1 + a2x2 + · · ·+ anxn = α,
where α, a1, a2, . . . , an ∈ R.
A linear inequality constraint is any inequality of the form
a1x1 + a2x2 + · · ·+ anxn ≤ α,
ora1x1 + a2x2 + · · ·+ anxn ≥ α,
where α, a1, a2, . . . , an ∈ R.
Lecture 2: Introduction to Linear Programming Linear Programming 6 / 46
![Page 17: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/17.jpg)
A linear function of the variables x1, x2, . . . , xn is any function f of the form
f (x) = c1x1 + c2x2 + · · ·+ cnxn
for fixed ci ∈ R i = 1, . . . , n.
A linear equality constraint is any equation of the form
a1x1 + a2x2 + · · ·+ anxn = α,
where α, a1, a2, . . . , an ∈ R.
A linear inequality constraint is any inequality of the form
a1x1 + a2x2 + · · ·+ anxn ≤ α,
ora1x1 + a2x2 + · · ·+ anxn ≥ α,
where α, a1, a2, . . . , an ∈ R.
Lecture 2: Introduction to Linear Programming Linear Programming 6 / 46
![Page 18: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/18.jpg)
A linear function of the variables x1, x2, . . . , xn is any function f of the form
f (x) = c1x1 + c2x2 + · · ·+ cnxn
for fixed ci ∈ R i = 1, . . . , n.
A linear equality constraint is any equation of the form
a1x1 + a2x2 + · · ·+ anxn = α,
where α, a1, a2, . . . , an ∈ R.
A linear inequality constraint is any inequality of the form
a1x1 + a2x2 + · · ·+ anxn ≤ α,
ora1x1 + a2x2 + · · ·+ anxn ≥ α,
where α, a1, a2, . . . , an ∈ R.
Lecture 2: Introduction to Linear Programming Linear Programming 6 / 46
![Page 19: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/19.jpg)
Compact Representation
maximize c1x1 + c2x2 + · · ·+ cnxn
subject to ai1xi + ai2x2 + · · ·+ ainxn ≤ αi i = 1, . . . , s
bi1xi + bi2x2 + · · ·+ binxn = βi i = 1, . . . , r .
Lecture 2: Introduction to Linear Programming Linear Programming 7 / 46
![Page 20: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/20.jpg)
Vector Inequalities: Componentwise
Let x , y ∈ Rn.
x =
x1x2...xn
y =
y1y2...yn
We write x ≤ y if and only if
xi ≤ yi , i = 1, 2, . . . , n .
Lecture 2: Introduction to Linear Programming Linear Programming 8 / 46
![Page 21: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/21.jpg)
Matrix Notation
c1x1 + c2x2 + · · ·+ cnxn = cT x
c =
c1c2...cn
x =
x1x2...xn
Lecture 2: Introduction to Linear Programming Linear Programming 9 / 46
![Page 22: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/22.jpg)
Matrix Notation
ai1xi + ai2x2 + · · ·+ ainxn ≤ αi i = 1, . . . , s
⇐⇒Ax ≤ a
A =
a11 a12 . . . a1na21 a22 . . . a2n
......
. . ....
as1 as2 . . . asn
a =
α1
α2
...αs
Lecture 2: Introduction to Linear Programming Linear Programming 10 / 46
![Page 23: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/23.jpg)
Matrix Notation
bi1xi + bi2x2 + · · ·+ binxn = βi i = 1, . . . , r
⇐⇒Bx = b
B =
b11 b12 . . . b1nb21 b22 . . . b2n
......
. . ....
br1 br2 . . . brn
b =
β1β2...βr
Lecture 2: Introduction to Linear Programming Linear Programming 11 / 46
![Page 24: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/24.jpg)
LP’s Matrix Notation
maximize cT xsubject to Ax ≤ a and Bx = b
c =
c1...cn
, a =
α1
...αs
, b =
β1...βr
.
A =
a11 . . . a1n. . .
as1 . . . asn
, B =
b11 . . . b1n. . .
br1 . . . brn
Lecture 2: Introduction to Linear Programming Linear Programming 12 / 46
![Page 25: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/25.jpg)
LP’s Matrix Notation
maximize cT xsubject to Ax ≤ a and Bx = b
c =
c1...cn
, a =
α1
...αs
, b =
β1...βr
.
A =
a11 . . . a1n. . .
as1 . . . asn
, B =
b11 . . . b1n. . .
br1 . . . brn
Lecture 2: Introduction to Linear Programming Linear Programming 12 / 46
![Page 26: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/26.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 27: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/27.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 28: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/28.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 29: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/29.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 30: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/30.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 31: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/31.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 32: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/32.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 33: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/33.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 34: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/34.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 35: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/35.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 36: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/36.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 37: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/37.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 38: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/38.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 39: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/39.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 40: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/40.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 41: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/41.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 42: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/42.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 43: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/43.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 44: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/44.jpg)
Applications of Linear Programing
A very short list:
resource allocation
production scheduling
warehousing
layout design
transportation scheduling
facility location
supply chain management
Model selection
Machine Learning
Compressed sensing
flight crew scheduling
portfolio optimization
cash flow matching
currency exchange arbitrage
crop scheduling
diet balancing
parameter estimation
. . .
Lecture 2: Introduction to Linear Programming Linear Programming 13 / 46
![Page 45: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/45.jpg)
Example: Plastic Cup Factory
A local family-owned plastic cup manufacturer wants to optimize their productionmix in order to maximize their profit. They produce personalized beer mugs andchampagne glasses. The profit on a case of beer mugs is $25 while the profit on acase of champagne glasses is $20. The cups are manufactured with a machinecalled a plastic extruder which feeds on plastic resins. Each case of beer mugsrequires 20 lbs. of plastic resins to produce while champagne glasses require 12lbs. per case. The daily supply of plastic resins is limited to at most 1800 pounds.About 15 cases of either product can be produced per hour. At the moment thefamily wants to limit their work day to 8 hours.
Model this problem as an LP.
Lecture 2: Introduction to Linear Programming Linear Programming 14 / 46
![Page 46: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/46.jpg)
Example: Plastic Cup Factory
A local family-owned plastic cup manufacturer wants to optimize their productionmix in order to maximize their profit. They produce personalized beer mugs andchampagne glasses. The profit on a case of beer mugs is $25 while the profit on acase of champagne glasses is $20. The cups are manufactured with a machinecalled a plastic extruder which feeds on plastic resins. Each case of beer mugsrequires 20 lbs. of plastic resins to produce while champagne glasses require 12lbs. per case. The daily supply of plastic resins is limited to at most 1800 pounds.About 15 cases of either product can be produced per hour. At the moment thefamily wants to limit their work day to 8 hours.
Model this problem as an LP.
Lecture 2: Introduction to Linear Programming Linear Programming 14 / 46
![Page 47: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/47.jpg)
LP Modeling
The four basic steps of LP modeling.
1 Identify and label the decision variables.
2 Determine the objective and use the decision variables to write an expressionfor the objective function as a linear function of the decision variables.
3 Determine the explicit constraints and write a functional expression for eachof them as a linear equation/inequality in the decision variables.
4 Determine the implicit constraints and write them as a linearequation/inequality in the decision variables.
Lecture 2: Introduction to Linear Programming Linear Programming 15 / 46
![Page 48: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/48.jpg)
LP Modeling
The four basic steps of LP modeling.
1 Identify and label the decision variables.
2 Determine the objective and use the decision variables to write an expressionfor the objective function as a linear function of the decision variables.
3 Determine the explicit constraints and write a functional expression for eachof them as a linear equation/inequality in the decision variables.
4 Determine the implicit constraints and write them as a linearequation/inequality in the decision variables.
Lecture 2: Introduction to Linear Programming Linear Programming 15 / 46
![Page 49: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/49.jpg)
LP Modeling
The four basic steps of LP modeling.
1 Identify and label the decision variables.
2 Determine the objective and use the decision variables to write an expressionfor the objective function as a linear function of the decision variables.
3 Determine the explicit constraints and write a functional expression for eachof them as a linear equation/inequality in the decision variables.
4 Determine the implicit constraints and write them as a linearequation/inequality in the decision variables.
Lecture 2: Introduction to Linear Programming Linear Programming 15 / 46
![Page 50: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/50.jpg)
LP Modeling
The four basic steps of LP modeling.
1 Identify and label the decision variables.
2 Determine the objective and use the decision variables to write an expressionfor the objective function as a linear function of the decision variables.
3 Determine the explicit constraints and write a functional expression for eachof them as a linear equation/inequality in the decision variables.
4 Determine the implicit constraints and write them as a linearequation/inequality in the decision variables.
Lecture 2: Introduction to Linear Programming Linear Programming 15 / 46
![Page 51: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/51.jpg)
LP Modeling
The four basic steps of LP modeling.
1 Identify and label the decision variables.
2 Determine the objective and use the decision variables to write an expressionfor the objective function as a linear function of the decision variables.
3 Determine the explicit constraints and write a functional expression for eachof them as a linear equation/inequality in the decision variables.
4 Determine the implicit constraints and write them as a linearequation/inequality in the decision variables.
Lecture 2: Introduction to Linear Programming Linear Programming 15 / 46
![Page 52: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/52.jpg)
Decision Variables
Determining the decision variables is the most difficult part of modeling.
To determining these variables it is helpful to put yourself in the shoes of thedecision maker, then ask What must he or she know to do their job.
In the real world, the modeler often follows the decision maker around for days,weeks, even months at a time recording all of the actions and decisions thisperson must make.
In this phase of modeling, it is very important to resist the temptation to makeassumptions about the nature of the solution.This last point cannot be over emphasized. Even the most experienced modelersoccasionally fall into this trap since such assumptions can enter in very subtleways.
Lecture 2: Introduction to Linear Programming Linear Programming 16 / 46
![Page 53: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/53.jpg)
Decision Variables
Determining the decision variables is the most difficult part of modeling.
To determining these variables it is helpful to put yourself in the shoes of thedecision maker, then ask What must he or she know to do their job.
In the real world, the modeler often follows the decision maker around for days,weeks, even months at a time recording all of the actions and decisions thisperson must make.
In this phase of modeling, it is very important to resist the temptation to makeassumptions about the nature of the solution.This last point cannot be over emphasized. Even the most experienced modelersoccasionally fall into this trap since such assumptions can enter in very subtleways.
Lecture 2: Introduction to Linear Programming Linear Programming 16 / 46
![Page 54: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/54.jpg)
Decision Variables
Determining the decision variables is the most difficult part of modeling.
To determining these variables it is helpful to put yourself in the shoes of thedecision maker, then ask What must he or she know to do their job.
In the real world, the modeler often follows the decision maker around for days,weeks, even months at a time recording all of the actions and decisions thisperson must make.
In this phase of modeling, it is very important to resist the temptation to makeassumptions about the nature of the solution.This last point cannot be over emphasized. Even the most experienced modelersoccasionally fall into this trap since such assumptions can enter in very subtleways.
Lecture 2: Introduction to Linear Programming Linear Programming 16 / 46
![Page 55: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/55.jpg)
Decision Variables
Determining the decision variables is the most difficult part of modeling.
To determining these variables it is helpful to put yourself in the shoes of thedecision maker, then ask What must he or she know to do their job.
In the real world, the modeler often follows the decision maker around for days,weeks, even months at a time recording all of the actions and decisions thisperson must make.
In this phase of modeling, it is very important to resist the temptation to makeassumptions about the nature of the solution.
This last point cannot be over emphasized. Even the most experienced modelersoccasionally fall into this trap since such assumptions can enter in very subtleways.
Lecture 2: Introduction to Linear Programming Linear Programming 16 / 46
![Page 56: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/56.jpg)
Decision Variables
Determining the decision variables is the most difficult part of modeling.
To determining these variables it is helpful to put yourself in the shoes of thedecision maker, then ask What must he or she know to do their job.
In the real world, the modeler often follows the decision maker around for days,weeks, even months at a time recording all of the actions and decisions thisperson must make.
In this phase of modeling, it is very important to resist the temptation to makeassumptions about the nature of the solution.This last point cannot be over emphasized. Even the most experienced modelersoccasionally fall into this trap since such assumptions can enter in very subtleways.
Lecture 2: Introduction to Linear Programming Linear Programming 16 / 46
![Page 57: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/57.jpg)
Decision Variables
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
B = number of cases of beer mugs produced dailyC = number of cases of champagne glasses produced daily
Lecture 2: Introduction to Linear Programming Linear Programming 17 / 46
![Page 58: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/58.jpg)
Decision Variables
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
B = number of cases of beer mugs produced daily
C = number of cases of champagne glasses produced daily
Lecture 2: Introduction to Linear Programming Linear Programming 17 / 46
![Page 59: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/59.jpg)
Decision Variables
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
B = number of cases of beer mugs produced dailyC = number of cases of champagne glasses produced daily
Lecture 2: Introduction to Linear Programming Linear Programming 17 / 46
![Page 60: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/60.jpg)
Objective Function
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Maximize Profit: Profit = Revenue − Costs
Profit = 25B + 20C
Lecture 2: Introduction to Linear Programming Linear Programming 18 / 46
![Page 61: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/61.jpg)
Objective Function
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Maximize Profit:
Profit = Revenue − Costs
Profit = 25B + 20C
Lecture 2: Introduction to Linear Programming Linear Programming 18 / 46
![Page 62: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/62.jpg)
Objective Function
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Maximize Profit: Profit = Revenue − Costs
Profit = 25B + 20C
Lecture 2: Introduction to Linear Programming Linear Programming 18 / 46
![Page 63: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/63.jpg)
Objective Function
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Maximize Profit: Profit = Revenue − Costs
Profit = 25B + 20C
Lecture 2: Introduction to Linear Programming Linear Programming 18 / 46
![Page 64: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/64.jpg)
Explicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Resin: 20B + 12C ≤ 1800
Labor: B/15 + C/15 ≤ 8
Lecture 2: Introduction to Linear Programming Linear Programming 19 / 46
![Page 65: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/65.jpg)
Explicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Resin:
20B + 12C ≤ 1800
Labor: B/15 + C/15 ≤ 8
Lecture 2: Introduction to Linear Programming Linear Programming 19 / 46
![Page 66: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/66.jpg)
Explicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Resin: 20B + 12C ≤ 1800
Labor: B/15 + C/15 ≤ 8
Lecture 2: Introduction to Linear Programming Linear Programming 19 / 46
![Page 67: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/67.jpg)
Explicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Resin: 20B + 12C ≤ 1800
Labor:
B/15 + C/15 ≤ 8
Lecture 2: Introduction to Linear Programming Linear Programming 19 / 46
![Page 68: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/68.jpg)
Explicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Resin: 20B + 12C ≤ 1800
Labor: B/15 + C/15 ≤ 8
Lecture 2: Introduction to Linear Programming Linear Programming 19 / 46
![Page 69: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/69.jpg)
Implicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Implicit Constraints:The decision variables are non-negative: 0 ≤ B, 0 ≤ C
Lecture 2: Introduction to Linear Programming Linear Programming 20 / 46
![Page 70: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/70.jpg)
Implicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Implicit Constraints:The decision variables are non-negative: 0 ≤ B, 0 ≤ C
Lecture 2: Introduction to Linear Programming Linear Programming 20 / 46
![Page 71: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/71.jpg)
Implicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Implicit Constraints:The decision variables are non-negative:
0 ≤ B, 0 ≤ C
Lecture 2: Introduction to Linear Programming Linear Programming 20 / 46
![Page 72: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/72.jpg)
Implicit Constraints
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Implicit Constraints:The decision variables are non-negative: 0 ≤ B, 0 ≤ C
Lecture 2: Introduction to Linear Programming Linear Programming 20 / 46
![Page 73: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/73.jpg)
The Plastic Cup Factory LP Model
maximize 25B + 20C
subject to 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Lecture 2: Introduction to Linear Programming Linear Programming 21 / 46
![Page 74: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/74.jpg)
The Hardest Part of Modeling: Decision Variables
Once again, the first step in the modeling process, identification of the decisionvariables, is always the most difficult.
Never be afraid to add more decision variables either to clarify the model orto improve its flexibility. Modern LP software easily solves problems withthousands of variables on a laptop, tens of thousands of variables on a server, oreven tens of millions of variables on specialized hardware and networks. It is moreimportant to get a correct, easily interpretable, and flexible model then to providea compact minimalist model.
LP model solutions found in many texts fall into the trap of trying to provide themost compact minimalist model with the fewest possible variables and constraints.Do not repeat this error in developing your own models.
Lecture 2: Introduction to Linear Programming Linear Programming 22 / 46
![Page 75: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/75.jpg)
The Hardest Part of Modeling: Decision Variables
Once again, the first step in the modeling process, identification of the decisionvariables, is always the most difficult.
Never be afraid to add more decision variables either to clarify the model orto improve its flexibility. Modern LP software easily solves problems withthousands of variables on a laptop, tens of thousands of variables on a server, oreven tens of millions of variables on specialized hardware and networks. It is moreimportant to get a correct, easily interpretable, and flexible model then to providea compact minimalist model.
LP model solutions found in many texts fall into the trap of trying to provide themost compact minimalist model with the fewest possible variables and constraints.Do not repeat this error in developing your own models.
Lecture 2: Introduction to Linear Programming Linear Programming 22 / 46
![Page 76: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/76.jpg)
The Hardest Part of Modeling: Decision Variables
Once again, the first step in the modeling process, identification of the decisionvariables, is always the most difficult.
Never be afraid to add more decision variables either to clarify the model orto improve its flexibility. Modern LP software easily solves problems withthousands of variables on a laptop, tens of thousands of variables on a server, oreven tens of millions of variables on specialized hardware and networks. It is moreimportant to get a correct, easily interpretable, and flexible model then to providea compact minimalist model.
LP model solutions found in many texts fall into the trap of trying to provide themost compact minimalist model with the fewest possible variables and constraints.
Do not repeat this error in developing your own models.
Lecture 2: Introduction to Linear Programming Linear Programming 22 / 46
![Page 77: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/77.jpg)
The Hardest Part of Modeling: Decision Variables
Once again, the first step in the modeling process, identification of the decisionvariables, is always the most difficult.
Never be afraid to add more decision variables either to clarify the model orto improve its flexibility. Modern LP software easily solves problems withthousands of variables on a laptop, tens of thousands of variables on a server, oreven tens of millions of variables on specialized hardware and networks. It is moreimportant to get a correct, easily interpretable, and flexible model then to providea compact minimalist model.
LP model solutions found in many texts fall into the trap of trying to provide themost compact minimalist model with the fewest possible variables and constraints.Do not repeat this error in developing your own models.
Lecture 2: Introduction to Linear Programming Linear Programming 22 / 46
![Page 78: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/78.jpg)
Graphical Solution of 2D LPs
We now graphically solve the LP model for the Plastic Cup Factory problem.
maximize 25B + 20C
subject to 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Lecture 2: Introduction to Linear Programming Linear Programming 23 / 46
![Page 79: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/79.jpg)
Graphical Solution of 2D LPs
14
15
13
12
1 4
11
10
9
8
5
6
7
4
3
2
1
2 3 5 6 7 8 9 10 11 12 13 14 15
feasible region
B
optimal value = $2625
20B + 12C = 1800
solution(BC
)=
(4575
)
115B + 1
15C = 8
C
objective normal n =(2520
)
1
Lecture 2: Introduction to Linear Programming Linear Programming 24 / 46
![Page 80: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/80.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 81: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/81.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 82: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/82.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 83: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/83.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 84: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/84.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.
Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 85: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/85.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).
Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 86: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/86.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 87: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/87.jpg)
Recap: Graphical Solution of 2D LPs
Step 1: Graph each of the linear constraints indicating on which side of theconstraint the feasible region must lie with an arrow. Don’t forget the implicitconstraints!
Step 2: Shade in the feasible region.
Step 3: Draw the gradient vector of the objective function.
Step 4: Place a straight-edge perpendicular to the gradient vector.Move the straight-edge in the direction of the gradient vector for maximization (orin the opposite direction for minimization).Move to the last point for which the straight-edge intersects the feasible region.
Step 5: The set of points of intersection between the straight-edge and thefeasible region is the set of solutions to the LP. Compute these points preciselyalong with the associated optimal value.
Lecture 2: Introduction to Linear Programming Linear Programming 25 / 46
![Page 88: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/88.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 89: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/89.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 90: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/90.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 91: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/91.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 92: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/92.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 93: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/93.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 94: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/94.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 95: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/95.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 96: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/96.jpg)
Sensitivity Analysis
Problems with the input data for real world LPs.
measurement error
changes over time
only an educated guess
model error
prospective studies
scenario analysis
LP approximates and nonlinear model/problem
We need to be able to study how the optimal value and solution change as theproblem input data change.
Lecture 2: Introduction to Linear Programming Linear Programming 26 / 46
![Page 97: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/97.jpg)
The Optimal Value Function
v(ε1, ε2) = maximize 25B + 20C
subject to 20B + 12C ≤ 1800 + ε1
115B + 1
15C ≤ 8 + ε2
0 ≤ B,C
Lecture 2: Introduction to Linear Programming Linear Programming 27 / 46
![Page 98: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/98.jpg)
Vertex Solutions
14
15
13
12
1 4
11
10
9
8
5
6
7
4
3
2
1
2 3 5 6 7 8 9 10 11 12 13 14 15
feasible region
B
optimal value = $2625
20B + 12C = 1800
solution(BC
)=
(4575
)
115B + 1
15C = 8
C
objective normal n =(2520
)
1
The optimal solution lies at a “corner point” or “vertex” of the feasible region.
Lecture 2: Introduction to Linear Programming Linear Programming 28 / 46
![Page 99: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/99.jpg)
Vertex Solutions
14
15
13
12
1 4
11
10
9
8
5
6
7
4
3
2
1
2 3 5 6 7 8 9 10 11 12 13 14 15
feasible region
B
optimal value = $2625
20B + 12C = 1800
solution(BC
)=
(4575
)
115B + 1
15C = 8
C
objective normal n =(2520
)
1
The optimal solution lies at a “corner point” or “vertex” of the feasible region.
Lecture 2: Introduction to Linear Programming Linear Programming 28 / 46
![Page 100: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/100.jpg)
Vertex Solutions
Conjecture: For a small range of perturbations to the resources, the vertexassociated with the current optimal solution moves but remains optimal.
v(ε1, ε2) = maximize 25B + 20Csubject to 20B + 12C ≤ 1800 + ε1
115B + 1
15C ≤ 8 + ε20 ≤ B,C
Lecture 2: Introduction to Linear Programming Linear Programming 29 / 46
![Page 101: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/101.jpg)
Vertex Solutions
The conjecture implies that the solution to the perturbed LP lies at theintersection of the two lines 20B + 12C = 1800 + ε1 and 1
15B + 115C = 8 + ε2 for
small values of ε1 and ε2; namely
B = 45− 45
2ε2 +
1
8ε1
C = 75 +75
2ε2 −
1
8ε1
v(ε1, ε2) = 25B + 20C = 2625 +375
2ε2 +
5
8ε1.
It can be verified by direct computation that this indeed yields the optimalsolution for small values of ε1 and ε2.
Lecture 2: Introduction to Linear Programming Linear Programming 30 / 46
![Page 102: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/102.jpg)
Vertex Solutions
The conjecture implies that the solution to the perturbed LP lies at theintersection of the two lines 20B + 12C = 1800 + ε1 and 1
15B + 115C = 8 + ε2 for
small values of ε1 and ε2; namely
B = 45− 45
2ε2 +
1
8ε1
C = 75 +75
2ε2 −
1
8ε1
v(ε1, ε2) = 25B + 20C = 2625 +375
2ε2 +
5
8ε1.
It can be verified by direct computation that this indeed yields the optimalsolution for small values of ε1 and ε2.
Lecture 2: Introduction to Linear Programming Linear Programming 30 / 46
![Page 103: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/103.jpg)
Vertex Solutions
The conjecture implies that the solution to the perturbed LP lies at theintersection of the two lines 20B + 12C = 1800 + ε1 and 1
15B + 115C = 8 + ε2 for
small values of ε1 and ε2; namely
B = 45− 45
2ε2 +
1
8ε1
C = 75 +75
2ε2 −
1
8ε1
v(ε1, ε2) = 25B + 20C
= 2625 +375
2ε2 +
5
8ε1.
It can be verified by direct computation that this indeed yields the optimalsolution for small values of ε1 and ε2.
Lecture 2: Introduction to Linear Programming Linear Programming 30 / 46
![Page 104: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/104.jpg)
Vertex Solutions
The conjecture implies that the solution to the perturbed LP lies at theintersection of the two lines 20B + 12C = 1800 + ε1 and 1
15B + 115C = 8 + ε2 for
small values of ε1 and ε2; namely
B = 45− 45
2ε2 +
1
8ε1
C = 75 +75
2ε2 −
1
8ε1
v(ε1, ε2) = 25B + 20C = 2625 +375
2ε2 +
5
8ε1.
It can be verified by direct computation that this indeed yields the optimalsolution for small values of ε1 and ε2.
Lecture 2: Introduction to Linear Programming Linear Programming 30 / 46
![Page 105: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/105.jpg)
Vertex Solutions
The conjecture implies that the solution to the perturbed LP lies at theintersection of the two lines 20B + 12C = 1800 + ε1 and 1
15B + 115C = 8 + ε2 for
small values of ε1 and ε2; namely
B = 45− 45
2ε2 +
1
8ε1
C = 75 +75
2ε2 −
1
8ε1
v(ε1, ε2) = 25B + 20C = 2625 +375
2ε2 +
5
8ε1.
It can be verified by direct computation that this indeed yields the optimalsolution for small values of ε1 and ε2.
Lecture 2: Introduction to Linear Programming Linear Programming 30 / 46
![Page 106: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/106.jpg)
Differentiability of the Optimal Value Function!
The optimal value function is differentiable for small values of ε1 and ε2.
v(ε1, ε2) = 2625 +375
2ε2 +
5
8ε1
∇v(ε1, ε2) =
58
3752
The components of the gradient are called the marginal values for the resources.
Lecture 2: Introduction to Linear Programming Linear Programming 31 / 46
![Page 107: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/107.jpg)
Differentiability of the Optimal Value Function!
The optimal value function is differentiable for small values of ε1 and ε2.
v(ε1, ε2) = 2625 +375
2ε2 +
5
8ε1
∇v(ε1, ε2) =
58
3752
The components of the gradient are called the marginal values for the resources.
Lecture 2: Introduction to Linear Programming Linear Programming 31 / 46
![Page 108: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/108.jpg)
Differentiability of the Optimal Value Function!
The optimal value function is differentiable for small values of ε1 and ε2.
v(ε1, ε2) = 2625 +375
2ε2 +
5
8ε1
∇v(ε1, ε2) =
58
3752
The components of the gradient are called the marginal values for the resources.
Lecture 2: Introduction to Linear Programming Linear Programming 31 / 46
![Page 109: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/109.jpg)
Differentiability of the Optimal Value Function!
The optimal value function is differentiable for small values of ε1 and ε2.
v(ε1, ε2) = 2625 +375
2ε2 +
5
8ε1
∇v(ε1, ε2) =
58
3752
The components of the gradient are called the marginal values for the resources.
Lecture 2: Introduction to Linear Programming Linear Programming 31 / 46
![Page 110: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/110.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 111: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/111.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 112: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/112.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 113: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/113.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 114: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/114.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 115: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/115.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 116: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/116.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 117: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/117.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 118: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/118.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 119: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/119.jpg)
The Theory of Linear Economic Models
Linear theory of production
John von Neumann, 1903-1957
Physics, Math, CS, Econ, Stats
Oskar Morgenstern, 1902-1977
Econ (Game Theory)
George Danzig, 1914-2005
Stats, Math, CS, Econ
Leonid Kantorovich, 1912-1986, Nobel Prize 1975
Math, Physics, Econ
Tjalling Koopmans, 1910-1985, Nobel Prize 1975
Econ, Physics, Math, Stats
Lecture 2: Introduction to Linear Programming Linear Programming 32 / 46
![Page 120: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/120.jpg)
The Production Model
in
materials
rawproducts
out
The production
process
The products are the raw materials reconfigured to look different.Profit is the difference between the purchase price of the raw materials and thesale price of their reconfigured form as products.Making a profit means that you sell the raw materials for more than you paid forthem.On a per unit basis, by how much does the production process increase the valueof the raw materials?
Lecture 2: Introduction to Linear Programming Linear Programming 33 / 46
![Page 121: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/121.jpg)
The Production Model
in
materials
rawproducts
out
The production
process
The products are the raw materials reconfigured to look different.
Profit is the difference between the purchase price of the raw materials and thesale price of their reconfigured form as products.Making a profit means that you sell the raw materials for more than you paid forthem.On a per unit basis, by how much does the production process increase the valueof the raw materials?
Lecture 2: Introduction to Linear Programming Linear Programming 33 / 46
![Page 122: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/122.jpg)
The Production Model
in
materials
rawproducts
out
The production
process
The products are the raw materials reconfigured to look different.Profit is the difference between the purchase price of the raw materials and thesale price of their reconfigured form as products.
Making a profit means that you sell the raw materials for more than you paid forthem.On a per unit basis, by how much does the production process increase the valueof the raw materials?
Lecture 2: Introduction to Linear Programming Linear Programming 33 / 46
![Page 123: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/123.jpg)
The Production Model
in
materials
rawproducts
out
The production
process
The products are the raw materials reconfigured to look different.Profit is the difference between the purchase price of the raw materials and thesale price of their reconfigured form as products.Making a profit means that you sell the raw materials for more than you paid forthem.
On a per unit basis, by how much does the production process increase the valueof the raw materials?
Lecture 2: Introduction to Linear Programming Linear Programming 33 / 46
![Page 124: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/124.jpg)
The Production Model
in
materials
rawproducts
out
The production
process
The products are the raw materials reconfigured to look different.Profit is the difference between the purchase price of the raw materials and thesale price of their reconfigured form as products.Making a profit means that you sell the raw materials for more than you paid forthem.On a per unit basis, by how much does the production process increase the valueof the raw materials?
Lecture 2: Introduction to Linear Programming Linear Programming 33 / 46
![Page 125: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/125.jpg)
The Optimal Value Function and Marginal Values
On a per unit basis, by how much does the production process increase the valueof the raw materials?
v(ε1, ε2) = maximize 25B + 20Csubject to 20B + 12C ≤ 1800 + ε1
115B + 1
15C ≤ 8 + ε20 ≤ B,C
Solution: The marginal values!
∇v(ε1, ε2) =
[5/8
375/2
]
Lecture 2: Introduction to Linear Programming Linear Programming 34 / 46
![Page 126: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/126.jpg)
The Optimal Value Function and Marginal Values
On a per unit basis, by how much does the production process increase the valueof the raw materials?
v(ε1, ε2) = maximize 25B + 20Csubject to 20B + 12C ≤ 1800 + ε1
115B + 1
15C ≤ 8 + ε20 ≤ B,C
Solution: The marginal values!
∇v(ε1, ε2) =
[5/8
375/2
]
Lecture 2: Introduction to Linear Programming Linear Programming 34 / 46
![Page 127: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/127.jpg)
The Optimal Value Function and Marginal Values
On a per unit basis, by how much does the production process increase the valueof the raw materials?
v(ε1, ε2) = maximize 25B + 20Csubject to 20B + 12C ≤ 1800 + ε1
115B + 1
15C ≤ 8 + ε20 ≤ B,C
Solution: The marginal values!
∇v(ε1, ε2) =
[5/8
375/2
]
Lecture 2: Introduction to Linear Programming Linear Programming 34 / 46
![Page 128: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/128.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 129: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/129.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 130: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/130.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 131: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/131.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 132: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/132.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 133: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/133.jpg)
Hidden Hand of the Market Place: Duality
In the market place there is competition for raw materials, or the inputs toproduction. This collective competition is the hidden hand that sets the price forgoods in the market place.
Is there a mathematical model for how these prices are set?
Let us think of the market as a separate agent in the market place. It is the agentthat owns and sells the raw materials of production.
The goal of the market is to make the most money possible from its resources bysetting the highest prices possible for them.
The market does not want to put the producers out of business, it just wants totake all of their profit.
How can we model this mathematically?
Lecture 2: Introduction to Linear Programming Linear Programming 35 / 46
![Page 134: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/134.jpg)
Hidden Hand of the Market Place: Duality
We answer this question in the context of the Plastic Cup Factory.
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Lecture 2: Introduction to Linear Programming Linear Programming 36 / 46
![Page 135: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/135.jpg)
Hidden Hand of the Market Place: Duality
We answer this question in the context of the Plastic Cup Factory.
A local family-owned plastic cup manufacturer wants to optimize their production mix in
order to maximize their profit. They produce personalized beer mugs and champagne
glasses. The profit on a case of beer mugs is $25 while the profit on a case of
champagne glasses is $20. The cups are manufactured with a machine called a plastic
extruder which feeds on plastic resins. Each case of beer mugs requires 20 lbs. of plastic
resins to produce while champagne glasses require 12 lbs. per case. The daily supply of
plastic resins is limited to at most 1800 pounds. About 15 cases of either product can be
produced per hour. At the moment the family wants to limit their work day to 8 hours.
Lecture 2: Introduction to Linear Programming Linear Programming 36 / 46
![Page 136: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/136.jpg)
Hidden Hand of the Market Place: Duality
By how much should the market increase the sale price of plastic resin and hourlylabor in order to wipe out the profit for the Plastic Cup Factory?
Define
0 ≤ y1 = price increase for a pound of resin
0 ≤ y2 = price increase for an hour of labor
These price increases should wipe out the per unit profitability for cases of bothbeer mugs and champagne glasses.
Lecture 2: Introduction to Linear Programming Linear Programming 37 / 46
![Page 137: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/137.jpg)
Hidden Hand of the Market Place: Duality
By how much should the market increase the sale price of plastic resin and hourlylabor in order to wipe out the profit for the Plastic Cup Factory?
Define
0 ≤ y1 = price increase for a pound of resin
0 ≤ y2 = price increase for an hour of labor
These price increases should wipe out the per unit profitability for cases of bothbeer mugs and champagne glasses.
Lecture 2: Introduction to Linear Programming Linear Programming 37 / 46
![Page 138: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/138.jpg)
Hidden Hand of the Market Place: Duality
By how much should the market increase the sale price of plastic resin and hourlylabor in order to wipe out the profit for the Plastic Cup Factory?
Define
0 ≤ y1 = price increase for a pound of resin
0 ≤ y2 = price increase for an hour of labor
These price increases should wipe out the per unit profitability for cases of bothbeer mugs and champagne glasses.
Lecture 2: Introduction to Linear Programming Linear Programming 37 / 46
![Page 139: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/139.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 140: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/140.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase =
20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 141: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/141.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2
≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 142: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/142.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 143: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/143.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses:
cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 144: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/144.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase =
12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 145: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/145.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2
≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 146: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/146.jpg)
Hidden Hand of the Market Place: Duality
production cost increase ≥ current profit
current profit
Beer Mugs: cost increase = 20y1 + 115y2 ≥ 25
Champagne Glasses: cost increase = 12y1 + 115y2 ≥ 20
Lecture 2: Introduction to Linear Programming Linear Programming 38 / 46
![Page 147: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/147.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 148: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/148.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2
Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 149: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/149.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 150: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/150.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 151: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/151.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!
Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 152: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/152.jpg)
Hidden Hand of the Market Place: Duality
Now minimize the total increase in the cost of raw materials subject to wiping outthe producer’s profit. Hopefully this will keep the Plastic Cup Factory in business.
minimize 1800y1 + 8y2Rewriting the Market’s price increase problem we get
minimize 1800y1 + 8y2
subject to 20y1 + y2/15 ≥ 2512y1 + y2/15 ≥ 200 ≤ y1, y2
This is another linear program!Let us compare this LP with the original LP.
Lecture 2: Introduction to Linear Programming Linear Programming 39 / 46
![Page 153: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/153.jpg)
Linear Programming Duality
Primal: max 25B + 20C
s.t. 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Dual: min 1800y1 + 8y2
s.t. 20y1 + 115y2 ≥ 25
12y1 + 115y2 ≥ 20
0 ≤ y1, y2
Lecture 2: Introduction to Linear Programming Linear Programming 40 / 46
![Page 154: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/154.jpg)
Linear Programming Duality
Primal:
max 25B + 20C
s.t. 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Dual: min 1800y1 + 8y2
s.t. 20y1 + 115y2 ≥ 25
12y1 + 115y2 ≥ 20
0 ≤ y1, y2
Lecture 2: Introduction to Linear Programming Linear Programming 40 / 46
![Page 155: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/155.jpg)
Linear Programming Duality
Primal: max 25B + 20C
s.t. 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Dual: min 1800y1 + 8y2
s.t. 20y1 + 115y2 ≥ 25
12y1 + 115y2 ≥ 20
0 ≤ y1, y2
Lecture 2: Introduction to Linear Programming Linear Programming 40 / 46
![Page 156: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/156.jpg)
Linear Programming Duality
Primal: max 25B + 20C
s.t. 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Dual:
min 1800y1 + 8y2
s.t. 20y1 + 115y2 ≥ 25
12y1 + 115y2 ≥ 20
0 ≤ y1, y2
Lecture 2: Introduction to Linear Programming Linear Programming 40 / 46
![Page 157: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/157.jpg)
Linear Programming Duality
Primal: max 25B + 20C
s.t. 20B + 12C ≤ 1800
115B + 1
15C ≤ 8
0 ≤ B,C
Dual: min 1800y1 + 8y2
s.t. 20y1 + 115y2 ≥ 25
12y1 + 115y2 ≥ 20
0 ≤ y1, y2
Lecture 2: Introduction to Linear Programming Linear Programming 40 / 46
![Page 158: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/158.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 159: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/159.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 160: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/160.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 161: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/161.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 162: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/162.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!
And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 163: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/163.jpg)
What is the Solution to the Dual?
Recall the goal of the Market:
The market wants to make the most money possible from its resources by settingthe highest prices it can without driving the producers out of business.
in
materials
rawproducts
out
The production
process
The marginal values give the per unit increase in the value of the resources due tothe production process.
The marginal values should be the solution to the dual!And indeed, they are the solution!
Lecture 2: Introduction to Linear Programming Linear Programming 41 / 46
![Page 164: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/164.jpg)
Linear Programming Duality: Matrix Notation
P
Primal: max cT x
s.t. Ax ≤ b
0 ≤ x
D
Dual: min bT y
s.t. AT y ≥ c
0 ≤ y
Lecture 2: Introduction to Linear Programming Linear Programming 42 / 46
![Page 165: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/165.jpg)
The Weak Duality Theorem of Linear Programming
Theorem: [Weak Duality Theorem]
If x ∈ Rn is feasible for P and y ∈ Rm is feasible for D, then
cT x ≤ yTAx ≤ bT y .
Thus, if P is unbounded, then D is necessarily infeasible, and if D is unbounded,then P is necessarily infeasible.
Moreover, if cT x = bT y with x feasible for P and y feasible for D, then x mustsolve P and y must solve D.
Lecture 2: Introduction to Linear Programming Linear Programming 43 / 46
![Page 166: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/166.jpg)
The Weak Duality Theorem of Linear Programming
Theorem: [Weak Duality Theorem]
If x ∈ Rn is feasible for P and y ∈ Rm is feasible for D, then
cT x ≤ yTAx ≤ bT y .
Thus, if P is unbounded, then D is necessarily infeasible, and if D is unbounded,then P is necessarily infeasible.
Moreover, if cT x = bT y with x feasible for P and y feasible for D, then x mustsolve P and y must solve D.
Lecture 2: Introduction to Linear Programming Linear Programming 43 / 46
![Page 167: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/167.jpg)
The Weak Duality Theorem of Linear Programming
Theorem: [Weak Duality Theorem]
If x ∈ Rn is feasible for P and y ∈ Rm is feasible for D, then
cT x ≤ yTAx ≤ bT y .
Thus, if P is unbounded, then D is necessarily infeasible, and if D is unbounded,then P is necessarily infeasible.
Moreover, if cT x = bT y with x feasible for P and y feasible for D, then x mustsolve P and y must solve D.
Lecture 2: Introduction to Linear Programming Linear Programming 43 / 46
![Page 168: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/168.jpg)
The Weak Duality Theorem of Linear Programming
Theorem: [Weak Duality Theorem]
If x ∈ Rn is feasible for P and y ∈ Rm is feasible for D, then
cT x ≤ yTAx ≤ bT y .
Thus, if P is unbounded, then D is necessarily infeasible, and if D is unbounded,then P is necessarily infeasible.
Moreover, if cT x = bT y with x feasible for P and y feasible for D, then x mustsolve P and y must solve D.
Lecture 2: Introduction to Linear Programming Linear Programming 43 / 46
![Page 169: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/169.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 170: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/170.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 171: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/171.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 172: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/172.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 173: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/173.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 174: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/174.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 175: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/175.jpg)
The Weak Duality Theorem of Linear Programming
Proof:
cT x =n∑
j=1
cjxj
≤n∑
j=1
(m∑i=1
aijyi )xj [0 ≤ xj , cj ≤m∑i=1
aijyi ⇒ cjxj ≤ (m∑i=1
aijyi )xj ]
= yTAx
=m∑i=1
(n∑
j=1
aijxj)yi
≤m∑i=1
biyi [0 ≤ yi ,n∑
j=1
aijxj ≤ bi ⇒ (n∑
j=1
aijxj)yi ≤ biyi ]
= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 44 / 46
![Page 176: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/176.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8
,
0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 177: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/177.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]
Dual: min 1800y1 + 8y2s.t. 20y1 + (1/15)y2 ≥ 25
12y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8
,
0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 178: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/178.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8
,
0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 179: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/179.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8
,
0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 180: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/180.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8,
0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 181: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/181.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2,
20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 182: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/182.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25,
12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 183: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/183.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 184: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/184.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 185: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/185.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625
= 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 186: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/186.jpg)
Test the WDT on the Plastic Cup Factory
Optimal Solution =
[4575
]
Marginal Values =
[5/8
375/2
]Dual: min 1800y1 + 8y2
s.t. 20y1 + (1/15)y2 ≥ 2512y1 + (1/15)y2 ≥ 200 ≤ y1, y2
Dual feasibility of the marginal values:
0 ≤ 5
8, 0 ≤ 375
2, 20 · 5
8+
1
15· 375
2≥ 25, 12 · 5
8+
1
15· 375
2≥ 20
Equivalence of primal-dual objectives (WDT):
cT x = 25 · 45 + 20 · 75 = 2625 = 1800 · 5
8+ 8 · 375
2= bT y
Lecture 2: Introduction to Linear Programming Linear Programming 45 / 46
![Page 187: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/187.jpg)
What the Weak Duality Theorem Does NOT Say
Infeasibility of either P or D does not imply the unboundedness of the other.
It is possible for both P and D to be infeasible.
Example:maximize 2x1 − x2
x1 − x2 ≤ 1−x1 + x2 ≤ −2
0 ≤ x1, x2
Lecture 2: Introduction to Linear Programming Linear Programming 46 / 46
![Page 188: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/188.jpg)
What the Weak Duality Theorem Does NOT Say
Infeasibility of either P or D does not imply the unboundedness of the other.
It is possible for both P and D to be infeasible.
Example:maximize 2x1 − x2
x1 − x2 ≤ 1−x1 + x2 ≤ −2
0 ≤ x1, x2
Lecture 2: Introduction to Linear Programming Linear Programming 46 / 46
![Page 189: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/189.jpg)
What the Weak Duality Theorem Does NOT Say
Infeasibility of either P or D does not imply the unboundedness of the other.
It is possible for both P and D to be infeasible.
Example:maximize 2x1 − x2
x1 − x2 ≤ 1−x1 + x2 ≤ −2
0 ≤ x1, x2
Lecture 2: Introduction to Linear Programming Linear Programming 46 / 46
![Page 190: Linear Programming - University of Washingtonburke/crs/407/lectures/L2-introduction.pdf · What is linear programming (LP)? A linear program is an optimization problem in nitely many](https://reader030.vdocuments.us/reader030/viewer/2022041204/5d545ca588c993e7228b68a5/html5/thumbnails/190.jpg)
What the Weak Duality Theorem Does NOT Say
Infeasibility of either P or D does not imply the unboundedness of the other.
It is possible for both P and D to be infeasible.
Example:maximize 2x1 − x2
x1 − x2 ≤ 1−x1 + x2 ≤ −2
0 ≤ x1, x2
Lecture 2: Introduction to Linear Programming Linear Programming 46 / 46