Fundamentals of Optimization Detailed
Fundamentals of Optimization Detailed
What is Optimization?
• Design civil engineering structures for minimum cost. • Optimal planning and scheduling.
• Design water-resource projects like dams to mitigate flood • Statistical analysis and models with minimum error.
damage while yielding maximum hydropower. • Optimal pipeline networks.
• Predict structural behavior by minimizing potential energy. • Inventory control.
CLASSIFICATION OF OPTIMIZATION
PROBLEMS
Classification Based on the Existence of Constraints
• Any optimization problem can be classified as constrained or unconstrained, depending on whether constraints exist
in the problem.
• In this category, the problem is to find values to a set of design parameters that make some prescribed function of these
parameters minimum subject to certain constraints.
• For example, the problem of minimum-weight design of a prismatic beam shown in Figure subject to a limitation on
the maximum deflection can be stated as follows:
where 𝜌 is the density and 𝛿tip is the tip deflection of the beam.
Such problems are called parameter or static optimization problems.
Classification Based on the Nature of the Design
Variables
• In the second category of problems, the objective is to find a set of design parameters, which are all continuous
functions of some other parameter, that minimizes an objective function subject to a set of constraints.
• If the cross-sectional dimensions of the rectangular beam are allowed to vary along its length as shown in Figure, the
optimization problem can be stated as:
• Nonlinear Programming Problem - If any of the functions among the objective and constraint functions in Eq. (1.1)
is nonlinear, the problem is called an NLP problem. This is the most general programming problem and all other
problems can be considered as special cases of the NLP problem.
• Geometric Programming Problem A GMP problem is one in which the objective function and constraints are
• expressed as posynomials in X
• A function h(X) is called a posynomial if h can be expressed as the sum of power terms each of the form
•
Classification Based on the Nature of the Equations
Involved
• Quadratic Programming Problem - A quadratic programming problem is a NLP problem with a quadratic objective
function and linear constraints.
• Linear Programming Problem - If the objective function and all the constraints in are linear functions of the design
variables, the mathematical programming problem is called a linear programming (LP) problem.
A manufacturing firm produces two products, A and B, using two limited resources. The maximum amounts of
resources 1 and 2 available per day are 1000 and 250 units, respectively. The production of 1 unit of product A
requires 1 unit of resource 1 and 0.2 unit of resource 2, and the production of 1 unit of product B requires 0.5 unit of
resource 1 and 0.5 unit of resource 2. The unit costs of resources 1 and 2 are given by the relations (0.375 -
0.00005u1) and (0.75 - 0.0001u2), respectively, where
ui denotes the number of units of resource i used (i = 1, 2). The selling prices per unit
of products A and B, pA and pB, are given by
where xA and xB indicate, respectively, the number of units of products A and B sold. Formulate the problem of
maximizing the profit assuming that the firm can sell all the units it manufactures.
Classification Based on the Permissible Values of the
Design Variables
• Depending on the values permitted for the design variables, optimization problems can be classified as integer and
real-valued programming problems.
• Integer Programming Problem If some or all of the design variables x1, x2, . . . ,xn of an optimization problem are
restricted to take on only integer (or discrete) values, the problem is called an integer programming problem .
• On the other hand, if all the design variables are permitted to take any real value, the optimization problem is called a
real-valued programming problem .
Classification Based on the Deterministic Nature of the
Variables
• Stochastic Programming Problem - A stochastic programming problem is an optimization problem in which some
or all of the parameters (design variables and/or preassigned parameters) are probabilistic (nondeterministic or
stochastic)
• Deterministic Programming Problem - refers to a computational problem where, given a specific set of input data,
the solution will always be the same, meaning the algorithm used to solve it will produce a predictable, unique output
every time it runs, with no element of randomness involved; essentially, the outcome is completely determined by the
input and the defined set of operations.
Classification Based on the Separability of the
Functions
• Optimization problems can be classified as separable and non-separable programming problems based on the
separability of the objective and constraint functions.
• Separable Programming Problem
– A function f (X) is said to be separable if it can be expressed as the sum of n single-variable functions, f1(x1), f2(x2), . . . , fn(xn)
– A separable programming problem is one in which the objective function and the constraints are separable .
• A retail store stocks and sells three different models of TV sets. The store cannot afford to have an inventory worth more
than $45 000 at any time. The TV sets are ordered in lots. It costs $a j for the store whenever a lot of TV model j is
ordered.
The cost of one TV set of model j is cj. The demand rate of TV model j is dj units per year. The rate at which the
inventory costs accumulate is known to be proportional to the investment in inventory at any time, with qj = 0.5, denoting
the constant of proportionality for TV model j. Each TV set occupies an area of sj = 0.40 m2 and the
maximum storage space available is 90 m2. The data known from the past experience are given below.
• Formulate the problem of minimizing the average annual cost of ordering and storing the TV sets.
Classification Based on the Number of Objective
Functions
• Depending on the number of objective functions to be minimized, optimization problems can be classified as single-
and multi-objective programming problems.
• Multi-objective Programming Problem A multi-objective programming problem can be stated as follows:
subject to
gj(X , j = 1, 2, . . . , m