Linear Optimization
Yinyu Ye
Department of Management Science and Engineering Stanford University
Stanford, CA 94305, U.S.A.
Introduction to Linear Optimization
The field of optimization is concerned with the study of maximization and minimization of mathematical functions. Very often the arguments of (i.e., variables or unknowns in) these functions are subject to side conditions or constraints. By virtue of its great utility in such diverse areas as applied science, engineering, economics, finance, medicine, and statistics, optimization holds an important place in the practical world and the scientific world. Indeed, as far back as the Eighteenth Century, the famous Swiss mathematician and physicist Leonhard Euler (1707-1783) proclaimeda that . . . nothing at all takes place in the Universe in which some rule of maximum or minimum does not appear.
aSee Leonhardo Eulero, Methodus Inviendi Lineas Curvas Maximi Minimive Proprietate Gaudentes,
Linear Programs and Extensions (in Standard Form)
Linear Programming
(LP )
minimizec
Tx
subject to
Ax = b,
x
≥ 0.
Linearly Constrained Optimization Problem
(LCOP )
minimizef (x)
subject to
Ax = b,
Linear Complementarity Problem Find nonnegative vectors
x
≥ 0, s ≥ 0
such that(LCP )
s = M x + q,
s
Tx = 0.
Rectified Linear Unit Operation:
s = max
{0, ay + b} ⇒ sx = 0, s = x + ay + b, (s, x) ≥ 0.
Conic Linear Programming
(CLP )
minimizec
• x
subject to
a
i• x = b
i, i = 1, ..., m,
x
∈ K,
whereK
is a closed convex cone.CLP: LP, SOCP, and SDP Examples
minimize2x
1+ x
2+ x
3 subject tox
1+ x
2+ x
3= 1,
(x
1; x
2; x
3)
≥ 0;
minimize2x
1+ x
2+ x
3 subject tox
1+ x
2+ x
3= 1,
√
x
22+ x
23≤ x
1.
minimize2x
1+ x
2+ x
3 subject tox
1+ x
2+ x
3= 1,
x
1x
2x
2x
3
≽ 0,
LP Terminology
•
decision variable/activity, data/parameter•
objective/goal/target, coefficient vector•
constraint/limitation/requirement, satisfied/violated•
equality/inequality constraint, direction of inequality, non-negativity•
constraint matrix/the right-hand side•
feasible/infeasible solution, interior feasible solution•
optimizers and optimum valuesLinear Programming Facts
•
The feasible region is a convex polyhedron.•
Every linear program is either feasible/bounded, feasible/unbounded, or infeasible.•
If feasible/bounded, every local optimizer is global and all optimizers form a convex polyhedron set.•
All optimizers are on the boundary of the feasible region.•
If the feasible region has an extreme point, then there must be an extreme optimizer.Linear Optimization Model and Formulation
•
Sort out data and parameters from the verbal description•
Define the set of decision variables•
Formulate the linear objective function of data and decision variablesTransportation Problem
min
∑
mi=1∑
nj=1c
ijx
ij s.t.∑
nj=1x
ij= s
i,
∀i = 1, ..., m
∑
m i=1x
ij= d
j,
∀j = 1, ..., n
x
ij≥ 0, ∀i, j.
2
3
n
1
2
m
.
.
.
.
.
.
s1
s2
sm
d2
d3
dn
Max-Flow Problem
Given a directed graph with nodes
1, ..., m
and edgesA
, where node1
is called source and nodem
is called the sink, and each edge(i, j)
has a flow rate capacityk
ij. The Max-Flow problem is to find the largest possible flow rate from source to sink.Let
x
ij be the flow rate from nodei
to nodej
. Then the problem can be formulated asmaximize
x
m1subject to
∑
j:(j,1)∈Ax
j1−
∑
j:(1,j)∈Ax
1j+ x
m1= 0,
∑
j:(j,i)∈A
x
ji−
∑
j:(i,j)∈A
x
ij= 0,
∀i = 2, ..., m − 1,
∑
j:(j,m)∈A
x
jm−
∑
j:(m,j)∈A
x
mj− x
m1= 0,
2
1
Sink
Source
6
3
3
4
5
3
4
4
2
3
7
3
Prediction Market I: World Cup Information Market
Order:#1
#2
#3
#4
#5
Argentina1
0
1
1
0
Brazil1
0
0
1
1
Italy1
0
1
1
0
Germany0
1
0
1
1
France0
0
1
0
0
Bidding Prize:π
0.75
0.35
0.4
0.95
0.75
Quantity limit:q
10
5
10
10
5
Order fill:x
x
1x
2x
3x
4x
5Prediction Market II: Call Auction Mechanism
Given
m
potential states that are mutually exclusive and exactly one of them will be realized at the maturity. An order is a bet on one or a combination of states, with a price limit (the maximum price the participant is willing to pay for one unit of the order) and a quantity limit (the maximum number of units or shares the participant is willing to accept).A contract on an order is a paper agreement so that on maturity it is worth a notional $
1
dollar if the order includes the winning state and worth $0
otherwise.Prediction Market III: Input Order Data
The
i
th order is given as(a
i·∈ R
m+, π
i∈ R
+, q
i∈ R
+)
:a
i· is the betting indication row vector where each component is either1
or0
a
i·= (a
i1, a
i2, ..., a
im)
where
1
is winning state and0
is non-winning state;π
i is the price limit for one unit of such a contract, andq
i is the maximum number of contract units the better like to buy.Prediction Market IV: Output Order-Fill Decisions
Let
x
i be the number of units or shares awarded to thei
th order. Then, thei
th bidder will pay the amountπ
i· x
i and the total amount collected would beπ
Tx =
∑
i
π
i· x
i.If the
j
th state is the winning state, then the auction organizer need to pay the winning bidders(
n∑
i=1a
ijx
i)
= a
T·jx
where column vector
a
·j= (a
1j; a
2j; ...; a
nj)
Prediction Market V: Worst-Case Profit Maximization
max
π
Tx
− max
j{a
T·jx
}
s.t.x
≤ q,
x
≥ 0.
max
π
Tx
− max(A
Tx)
s.t.x
≤ q,
x
≥ 0.
This is NOT a linear program.However, the problem can be rewritten as
max
π
Tx
− y
s.t.
A
Tx
− e · y ≤ 0,
x
≤ q,
x
≥ 0,
wheree
is the vector of all ones. This is a linear program.max
π
Tx
− y
s.t.