CBEg 6152-Process Systems Engineering (PSE)

Optimization of Process Systems

By
Dr. Eng. Shegaw Ahmed
School of Chemical & Bio Engineering
Addis Ababa Institute of Technology
Addis Ababa University

March, 2018
1
 SCHOOL OF CHEMICAL AND BIO ENGINEERING
ADDIS ABABA INSTITUTE OF TECHNOLOGY(AAiT)
ADDIS ABABA UNIVERSITY (AAU)
CBEg 6152-PSE-Outline

CHAPTER-1-Introduction to Process
CHAPTER-3-Introduction to Process Integration
Systems Engineering
• Definitions of Process Systems Engineering • Synthesis of Heat Exchange Networks
• The Concept of Systems Engineering • Synthesis of Mass Exchange Networks
• The Paradigms of Process Systems Engineering

CHAPTER-4-Optimization of Process Systems

• Application of Optimization Techniques for Design

CHAPTER-2-Process Modelling and Simulation and Operation
• Multivariable Search Techniques
• Mathematical Modelling Techniques of Chemical • Linear Programming (LP)
Processes • Nonlinear Programming (NLP)
• Conceptual Process Synthesis and Analysis • Dynamic Programming
• Fundamental of Process Control Systems • Integer (IP) and Mixed Integer Programming
• Process Engineering Fundamentals (MIP)
• Steady State and Dynamic Simulation • Distribution and Critical Path Networks
 Chapter 4. …Introduction to Optimization
• The chemical industry has undergone significant changes during the past few
decades years due to
o the increased cost of energy,

o increasingly stringent(rigorous & harsh) environmental regulations

o global competition in product pricing and quality

• One of the most important engineering tools for addressing these issues is
OPTIMIZATION

• OPTIMIZATION is the use of specific methods to determine the most cost-

effective and efficient solution to a problem or design for a process. This
technique is one of the major quantitative tools in industrial decision making

• Modifications in plant design and operating procedures have been

implemented to reduce costs and meet constraints, with an emphasis on
improving efficiency and increasing profitability
3
 Chapter 4. …What Optimization is all about?

• Typical problems in chemical engineering process design or plant

operation have many (possibly an infinite number) solutions.

• Optimization is concerned with selecting the best among the entire set
by efficient quantitative methods.

• Computers and associated software make the necessary

• computations feasible and cost-effective. Obtaining useful information

using computers, however, requires
 critical analysis of the process or design,

 insight about what the appropriate performance objectives are (i.e., what is to be
accomplished), and

 use of past experience, sometimes called engineering judgment

4
Chapter 4. …What Optimization is all about?

o want to find
• the conditions
• that give the BEST performance

BEST = MAXIMUM or MINIMUM

o solving for
 values of the variables
p e.g. temperature

 that yield the BEST value of the performance criterion

p e.g. max profit & yield

 subject to various constraints

p e.g. limitation on mass & energy, capacity limit

 using systematic quantitative methods

p e.g. numerical approach

5
Chapter 4. …What Optimization is all about?

• Why do we optimize in process design?

• What do we optimize in process design?

Objective(s)

• How do we optimize in process design?

Technique(s)
6
 Chapter 4. …Models as basis for optimization

o model is expressed in variables

o model of what we want to find the best value = objective
function
o model of restrictions in the system = constraints
o so, when optimizing/solving an optimization problem =
finding variables’ values that give the best value for
objective function

7
Chapter 4. …Building a model/Modelling

8
 Chapter 4. …General Formulation

Problem Statements and Terminologies

9
 Chapter 4. …General Formulation

 The main elements of any constrained optimization problem are: Objective

Function, Decision Variables, Constraints
 Objective Function: this is a mathematical expression that combines the
variables to express your goal. It may represent:
o economic objective: may be profits, costs of production etc
o Technical objective: may be the yield from the reactor that needs to be
maximized, minimum size of an equipment etc.
 Technical objectives are ultimately related to economics
 Decision Variables: the values of the variables are not known when you start
the problem. The variables usually represent things that you can adjust or
control, e.g. the rate at which to manufacture items. The goal is to find values of
the variables that provide the best value of the objective function.
 Constraints: these are mathematical expressions that combine the variables to
express the limits on the possible solutions. 10
 Chapter 4. …General Formulation

In order to find the solution of Optimization problems, it is important:

 Analyze the properties of its mathematical expressions
o find a formulation of the optimization problem,
 Analyze the mathematical structure of the problem, classify it according
to this structure and find appropriate methods for each of them
o using a suitable algorithm to solve the formulation

11
 Chapter 4. …Examples of applications of Optimization

Optimization can be applied in numerous ways to chemical processes and

plants.
 Determining the best sites for plant location
 Sizing and layout of a pipeline
 Designing equipment and an entire plant
 Scheduling maintenance and equipment replacement
 Operating equipment, such as tubular reactors, columns, and
absorbers, etc.
 Evaluating plant data to construct a model of a process
 Allocating resources or services among several processes
 Planning and scheduling construction
12
 Chapter 4. …Introduction to Optimization

Lets consider the optimization of a distillation tower, to build our ability

to formulate and solve optimization problems. We would like to
understand the new modeling concepts.

Energy cost, Product cost (value), Product purities

13
 Chapter 4. …Introduction to Optimization

What is the key characteristic of optimization problems?

Key characteristic
o A tradeoff exists between one or more variables and the objective function.
o We must identify the key tradeoffs before building mathematical models!
o We must understand the problem qualitatively, before we solve it quantitatively.

14
Chapter 4. …Optimization Examples

Economics: Plant Design

15
Chapter 4. …Optimization Examples

Economics: Company logistics and supply chain management

16
 Chapter 4. …Development Stages-Optimization Problem Formulation

Step 1: Problem Definition

Objectives
o goal of optimization
o performance measurement
o example : max profit, min energy cost, etc
o max{F(x, y)} = min{−F(x, y)}.
Process models
o assumptions & constants
o equations to represent relationships material & energy balances,
thermodynamics, kinetics, etc
o constraints and bounds to represent limitations operating restrictions, bounds &
limits
A point which satisfies the process model equations is a feasible point. Otherwise, it is
an infeasible point.
17
 Chapter 4. …Development Stages

Step 2: Problem Formulation

o Standard form
o Degrees of freedom analysis
 over-, under- or exactly specified,
 decision variables (independent vs. dependent)
o Scaling of variables units of measure, scaling factors
Step 3: Problem Solution
o Technique selection
 matched to problem type
 exploit problem structure
 knowledge of algorithms strengths & weaknesses
o Starting points . . . usually several and compare
o Algorithm tuning
 termination criteria; convergence tolerances; step length 18
 Chapter 4. …Development Stages

Step 4: Results Analysis

o Solution uniqueness
o Perturbation of optimization problem
 effects of assumptions
 variation (uncertainty) in problem parameters
 variation in prices/costs

Solution of an optimization
problem requires all of the steps

 full understanding is developed by following the complete cycle

 good decisions require a full understanding of the problem
 shortcuts lead to bad decisions.
19
 Chapter 4. …The importance of good formulation

o The conclusions are drawn from the model (mathematical program), not
from the problem!
o An inadequate model typically leads to false conclusions
o The model must be computationally tractable: its analysis must be practical!
20
 Chapter 4. …Example

A Transportation Problem
Let’s begin with a much simplified example of a problem that might arise in
manufacturing and transportation. A chemical company has 2 factories F1 and F2 and a
dozen retail outlets R1, R2, . . . , R12. Each factory Fi can produce ai tons of a certain
chemical product each week; ai is called the capacity of the plant. Each retail outlet Rj
has a known weekly demand of bj tons of the product. The cost of shipping one ton of the
product from factory Fi to retail outlet Rj is cij.

The problem is to determine how much of the product to ship from each factory to each
outlet so as to satisfy all the requirements and minimize cost. The variables of the
problem are xij , i = 1, 2, j = 1, 2, . . . , 12, where xij is the number of tons of the product
shipped from factory Fi to retail outlet Rj;

21
Chapter 4. …Example

22
 Chapter 4. …Optimization Model Formulation

Standard Model
Optimization models (also called mathematical programs)
represent problem choices as decision variables and seek values
that maximize (or minimize) objective functions of the decision
variables subject to constraints on variable values expressing the
limits on possible decision choices

23
 Chapter 4. …Decision Variables, x

o At the formulation stage, decision variables can be grouped into two

categories:
 Independent variables, whose values can be changed independently to
modify the behavior of the system
 Dependent variables, whose behavior is determined by the values selected
for the independent variables
Such a grouping helps understanding
o At the solution stage, independent and dependent variables need not be
distinguished: the solution method just sees an optimization problem
involving many variables
Examples of Decision Variables
 Design: reactor volume, number of trays, heat exchanger area, etc.
 Operations: temperature, flow, pressure, valve opening, etc.
 Management: feed type, purchase price, sales price, etc.

24
 Chapter 4. …Decision Variables, x

Discrete vs. Continuous Decision Variables

• A variable is discrete if it is limited to a fixed or countable set of values (e.g.,
all-or-nothing, either-or decisions)
• A variable is continuous if it can take on any value in a specified interval
Problems with both discrete and continuous variables are called mixed-integer
programs

Heuristics
When there is an option, modeling with continuous variables is preferred to discrete
variables

Variable Bounds, and Main constraints

• Variable bounds: specify the domain of definition for decision variables: the
set of values for which the variables have meaning
• Main constraints of optimization models: specify the restrictions and
interactions, other than variable bounds, that limit decision variable values
25
 Chapter 4. …Example

To clarify these ideas let us consider a simple example. A company manufactures four
products, requiring time on two machines and two types (skilled and unskilled) of labour. The
amount of machine time and labor (in hours) needed to produce a unit of each product and
the sales prices in dollars per unit of each product are given in the following table:

Each month, 700 hours are available on machine 1 and 500 hours on machine 2. Each
month, the company can purchase up to 600 hours of skilled labour at $8 per hour and up to
650 hours of unskilled labour at $6 per hour. The company wants to determine how much of
each product it should produce each month and how much labor to purchase in order to
maximize its profit (i.e., revenue from sales minus labour costs).

You need to determine the variables, the objective function, & the constraints
26
 Chapter 4. …Solution

The formulation
o We wish to find a formulation for this problem, i.e. we need to determine the
variables, the objective function, and the constraints
Variables
o The company must decide how much of each product to manufacture; we capture
this by introducing a variable xi for each i ∈ {1, 2, 3, 4} for the number of units of
product i to manufacture.
o As part of the planning process, the company must also decide on the number of
hours of skilled, ys, and unskilled, yu, labour that it wants to purchase.
Objective function
o Deciding on a production plan now amounts to finding values for variables x1, . . .
, x4, ys and yu. Once the values for these variables have been found, company’s
profit is easily expressed by the following function

27
Chapter 4. …Solution

28
 Chapter 4. …Unimodal and multi modal functions

Functions with only one extreme point, which is termed as

Unimodal otherwise it is termed as Multimodal

29
 Chapter 4. …Convexity and Concavity of function

To establish whether an optimal point is a Local or a Global Optimum, the

concepts of Convexity and Concavity is introduced.
If a straight line is drawn between any two points on the function, then the
function is Convex if all the values of this line lie above the curve
and if a straight line drawn between any two points on the function, then if all
values on this line lie below the curve, the function is Concave

30
 Chapter 4. …Analytical Techniques: Classical Optimization Techniques

o The classical methods of optimization are useful in finding the optimum

solution of continuous and differentiable functions
o These methods are analytical and make use of the techniques of differential
calculus in locating the optimum points
o However, some of the practical problems involve objective functions that are
not continuous and/or differentiable, the classical optimization techniques
have limited scope in practical applications.

31
Chapter 4. … Analytical Techniques-single variable optimization

32
 Chapter 4. … Analytical Techniques-single variable optimization

An example of a one-dimensional analytical optimization problem would be to

find the global minimum of

f objective fun, x independent variable. The global optimum x∗ of such a problem

are located at one of the three possible locations:
• At the stationary point(local max or local min)
• At the boundary formed by the constraints of the problem
• At the discontinuity in the objective function or the first derivative of the
function
To find the global min or max it is necessary to compare function values at
the stationary points, at boundaries, and at discontinuities.
Necessary and sufficient conditions

33
Chapter 4. …Analytical Techniques: Classical Optimization Techniques

34
Chapter 4. …Analytical Techniques: Classical Optimization Techniques

35
Chapter 4. …Example

36
Chapter 4. …Unconstrained Multivariable Optimization

39
Chapter 4. …Unconstrained Multivariable Optimization

42
Chapter 4. …Example

43
Chapter 4. … Unconstrained multivariable optimization problem

44
Chapter 4. … Unconstrained multivariable optimization problem

47
Chapter 4. …Unconstrained multivariable optimization problem

48
 Chapter 4. …Linear Programing (LP)

• LP is one of the most widely used optimization techniques and perhaps the most
effective
• It deals with the optimization (maximization or minimization) of linear objective
functions subject to linear constraints.
solution methods
o Graphical Solution
o Simplex Method
• LP problems involving only two variables can be effectively solved by a graphical
technique which provides a pictorial representation of the solution.
• step 1: formulate the given problem as a linear programming problem
• step 2: plot the given constraints as equalities on x1 − x2 coordinate plane and determine the
convex region formed by them
• step 3: determine the vertices of the convex region and find the value of objective function at
each vertex. The vertex which gives the optimal value of the objective function gives the desired
optimal solution to the problem. 49
Chapter 4. …Graphical Solution Example 1

Solution

50
Chapter 4. …Example 2

Solution

51
Chapter 4. …Example 3

Solution

52
Chapter 4. …Example 3

Solution

53
Chapter 4. …Standard form of LP

54
 Chapter 4. …Simplex Method

o For linear programming problems involving two variables, the

graphical solution method introduced before is convenient.

o However, for problems involving more than two variables or problems

involving a large number of constraints, it is better to use solution
methods that are adaptable to computers.

o One such method is called the simplex method

o It provides us with a systematic way of examining the vertices of the

feasible region to determine the optimal value of the objective
function.
55
 Chapter 4. …Simplex Method

o The basic LP problem consists of two major parts: the objective

function and a set of constraints. In Standard Form

56
Chapter 4. …Simplex Method

57
 Chapter 4. …Simplex Method

o Note that for a linear programming problem in standard form,

the objective function is to be maximized, not minimized

o A basic solution of a linear programming problem in standard

form is a solution ( x1, x2,…xn, s1, s2,…sm) of the constraint
equations in which at most m variables are nonzero––the
variables that are nonzero are called basic variables. A basic
solution for which all variables are nonnegative is called a basic
feasible solution.

58
 Chapter 4. …Simplex Method

The simplex method is carried out by performing elementary row operations on a

matrix that we call the simplex tableau. This tableau consists of the augmented
matrix corresponding to the constraint equations together with the coefficients of
the objective function

59
Chapter 4. …Simplex Method-Pivoting

60
 Chapter 4. …Simplex Method

The simplex method is carried out by performing elementary row operations on a

matrix that we call the simplex tableau. This tableau consists of the augmented
matrix corresponding to the constraint equations together with the coefficients of
the objective function

61
Chapter 4. …Simplex Method

62
 Chapter 4. …Simplex Method-Pivoting

Example: The Simplex Method with Three Decision Variables

63
Chapter 4. …Simplex Method-Pivoting

64
 Chapter 4. …Simplex Method-Example

A manufacturer produces three types of plastic fixtures. The time required

for molding, trimming, and packaging is given in the Table below. (Times
are given in hours per dozen fixtures.)

How many dozen of each type of fixture should be produced to obtain

a maximum profit?

65
Chapter 4. …Simplex Method-Example

66
Chapter 4. …Simplex Method-Example

67
Chapter 4. …Non-Linear Programming

68
 Chapter 4. …Non-Linear Programming

o Direct Substitution
o Lagrange multiplier method

69
 Chapter 4. …Non-Linear Programming-Direct Substitution

o One method of handling just one or two linear or nonlinear

equality constraints is to solve explicitly for one variable and
eliminate that variable from the problem formulation.
o This is done by Direct Substitution in the objective function and
constraint equations in the problem.
Example
Suppose you want to minimize the following objective function that is
subject to a single equality constraint

70
Chapter 4. …Non-Linear Programming-Direct Substitution

71
Chapter 4. …Non-Linear Programming-Lagrangian Multiplier Method

72
Chapter 4. …Non-Linear Programming-Lagrangian Multiplier Method

Example

73
Chapter 4. …Non-Linear Programming-Problems Containing only Equality
Constraints

74
Chapter 4. …Non-Linear Programming-Problems Containing only Equality
Constraints

75
Chapter 4. …Non-Linear Programming-Problems Containing only inequality
Constraints

76
Chapter 4. …Non-Linear Programming-Problems Containing only inequality
Constraints

77
Chapter 4. …Non-Linear Programming-Example

78
Chapter 4. …Non-Linear Programming-solution

79
Chapter 4. …Non-Linear Programming-Solution

80
Chapter 4. …Non-Linear Programming-Solution

81
 Chapter 4. …Non-Linear Programming-Examples

1. Find the dimensions of a box of largest volume that

can be inscribed in a sphere of unit radius

2. Find the optimal dimensions for a heated cylindrical

tank designed to hold 10 m3 of fluid. The ends and
sides cost $ 200/m2 and $ 100/m2, respectively. In
addition, a coating is applied to the entire tank area
at a cost of $ 50/m2.

82
Chapter 4. …Non-Linear Programming-Example

83
Chapter 4. …Non-Linear Programming-Example

84
Chapter 4. …Non-Linear Programming-Example

85
Chapter 4. …Non-Linear Programming-Exercise

86
Chapter 4. …Mixed Integer Programming-Reading assignment

87
Chapter 4. …Classification of the types of problems that are encountered in
optimization with discrete variables

88
Chapter 4. …-Some Integer-Programming Models

89
Chapter 4. …-Some Integer-Programming Models

90
Chapter 4. …-More Examples in Chemical Engineering

91
Chapter 4. …-More Examples in Chemical Engineering

92