Estimation of Project Metrics

Introduction
After gathering the entire requirements specific to software project usually we need to
think about different solution strategy for the project. Expert business analysts are
analyzing their benefits and as well as their shortcomings by means of cost, time and
resources require to develop it.
In this experiment, we will learn how to estimate cost, effort and duration for a software
project, and then select one solution approach which will be found suitable to fulfill the
organizational goal.
Objectives
After completing this experiment you will be able to:
 Categorize projects using COCOMO, and estimate effort and development time
required for a project
 Estimate the program complexity and effort required to recreate it using Halstead's
metrics
Time Required - Around 3.00 hours
Project Estimation Techniques
A software project is not just about writing a few hundred lines of source code to achieve a
particular objective. The scope of a software project is comparatively quite large, and such
a project could take several years to complete. However, the phrase "quite large" could only
give some (possibly vague) qualitative information. As in any other science and
engineering discipline, one would be interested to measure how complex a project is. One
of the major activities of the project planning phase, therefore, is to estimate various
project parameters in order to take proper decisions. Some important project parameters
that are estimated include:
 Project size: What would be the size of the code written say, in number of lines, files,
modules?
 Cost: How much would it cost to develop a software? A software may be just pieces
of code, but one has to pay to the managers, developers, and other project
personnel.
 Duration: How long would it be before the software is delivered to the clients?
 Effort: How much effort from the team members would be required to create the
software?
In this experiment we will focus on two methods for estimating project metrics: COCOMO
and Halstead's method.
COCOMO
COCOMO (Constructive Cost Model) was proposed by Boehm. According to him, there
could be three categories of software projects: organic, semidetached, and embedded. The
classification is done considering the characteristics of the software, the development team
and environment. These product classes typically correspond to application, utility and
system programs, respectively. Data processing programs could be considered as
application programs. Compilers, linkers, are examples of utility programs. Operating
systems, real-time system programs are examples of system programs. One could easily
apprehend that it would take much more time and effort to develop an OS than an
attendance management system.
The concept of organic, semidetached, and embedded systems are described below.
 Organic: A development project is said to be of organic type, if
 The project deals with developing a well understood application
 The development team is small
 The team members have prior experience in working with similar types of
projects
 Semidetached: A development project can be categorized as semidetached type, if
 The team consists of some experienced as well as inexperienced staff
 Team members may have some experience on the type of system to be
developed
 Embedded: Embedded type of development project are those, which
 Aims to develop a software strongly related to machine hardware
 Team size is usually large
Boehm suggested that estimation of project parameters should be done through three
stages: Basic COCOMO, Intermediate COCOMO, and Complete COCOMO.
Basic COCOMO Model
The basic COCOMO model helps to obtain a rough estimate of the project parameters.
It estimates effort and time required for development in the following way:
Effort = a * (KDSI)b PMTdev = 2.5 * (Effort)c Monthswhere
 KDSI is the estimated size of the software expressed in Kilo Delivered Source
Instructions
 a, b, c are constants determined by the category of software project
 Effort denotes the total effort required for the software development,
expressed in person months (PMs)
 Tdev denotes the estimated time required to develop the software (expressed
in months)
The value of the constants a, b, c are given below:

Software project a b c

Organic 2.4 1.05 0.38

Semi-detached 3.0 1.12 0.35

Embedded 3.6 1.20 0.32

Intermediate COCOMO Model

The basic COCOMO model considers that effort and development time depends only on
the size of the software. However, in real life there are many other project parameters
that influence the development process. The intermediate COCOMO take those other
factors into consideration by defining a set of 15 cost drivers (multipliers) as shown in
the table below [i]. Thus, any project that makes use of modern programming
practices would have lower estimates in terms of effort and cost. Each of the 15 such
attributes can be rated on a six-point scale ranging from "very low" to "extra high" in
their relative order of importance. Each attribute has an effort multiplier fixed as per
the rating. The product of effort multipliers of all the 15 attributes gives the Effort
Adjustment Factor (EAF).

Cost drivers for INtermediate COCOMO

(Source: http://en.wikipedia.org/wiki/COCOMO)

Ratings

Extra
Cost Drivers Very Low Low Nominal High Very High High

Product attributes

Required software
0.75 0.88 1.00 1.15 1.40
reliability

Size of application
0.94 1.00 1.08 1.16
database

Complexity of the
0.70 0.85 1.00 1.15 1.30 1.65
product

Hardware attributes

Run-time performance
1.00 1.11 1.30 1.66
constraints

Memory constraints 1.00 1.06 1.21 1.56

Volatility of the virtual

0.87 1.00 1.15 1.30
machine environment

Required turnabout time 0.87 1.00 1.07 1.15

Personnel attributes

Analyst capability 1.46 1.19 1.00 0.86 0.71

Applications experience 1.29 1.13 1.00 0.91 0.82

Software engineer
1.42 1.17 1.00 0.86 0.70
capability

Virtual machine
1.21 1.10 1.00 0.90
experience

Programming language
1.14 1.07 1.00 0.95
experience
 Cost drivers for INtermediate COCOMO
(Source: http://en.wikipedia.org/wiki/COCOMO)

Ratings

Extra
Cost Drivers Very Low Low Nominal High Very High High

Project attributes

Application of software
1.24 1.10 1.00 0.91 0.82
engineering methods

Use of software tools 1.24 1.10 1.00 0.91 0.83

Required development
1.23 1.08 1.00 1.04 1.10
schedule

EAF is used to refine the estimates obtained by basic COCOMO as follows: Effort|
corrected = Effort * EAFTdev|corrected = 2.5 * (Effort| corrected) c

Complete COCOMO Model

Both the basic and intermediate COCOMO models consider a software to be a single
homogeneous entity -- an assumption, which is rarely true. In fact, many real life
applications are made up of several smaller sub-systems. (One might not even develop
all the sub-systems -- just use the available services). The complete COCOMO model
takes these factors into account to provide a far more accurate estimate of project
metrics.
To illustrate this, consider a very popular distributed application: the ticket booking
system of the Indian Railways. There are computerized ticket counters in most of the
railway stations of our country. Tickets can be booked / cancelled from any such
counter. Reservations for future tickets, cancellation of reserved tickets could also be
performed. On a high level, the ticket booking system has three main components:
 Database
 Graphical User Interface (GUI)
 Networking facilities
Among these, development of the GUI is considered as an organic project type; the
database module could be considered as a semi-detached software. The networking
module can be considered as an embedded software. To obtain a realistic cost, one
should estimate the costs for each component separately, and then add it up.
Advantages of COCOMO
COCOMO is a simple model, and should help one to understand the concept of project
metrics estimation.
 Drawbacks of COCOMO
COCOMO uses KDSI, which is not a proper measure of a program's size. Indeed,
estimating the size of a software is a difficult task, and any slight miscalculation could
cause a large deviation in subsequent project estimates. Moreover, COCOMO was
proposed in 1981 keeping the waterfall model of project life cycle in mind [2]. It fails
to address other popular approaches like prototype, incremental, spiral, agile models.
Moreover, in present day a software project may not necessarily consist of coding of
every bit of functionality. Rather, existing software components are often used and
glued together towards the development of a new software. COCOMO is not suitable in
such cases.
COCOMO II was proposed later in 2000 to many of address these issues.
Halstead's Complexity Metrics
Halstead took a linguistic approach to determine the complexity of a program.
According to him, a computer program consists of a collection of different operands
and operators. The definition of operands and operators could, however, vary from
one person to another and one programming language to other. Operands are usually
the implementation variables or constants -- something upon which an operation
could be performed. Operators are those symbols that affects the value of operands.
Halstead's metrics are computed based on the operators and operands used in a
computer program. Any given program has the following four parameters:
 n1: Number of unique operators used in the program
 n2: Number of unique operands used in the program
 N1: Total number of operators used in the program
 N2: Total number of operands used in the program
Using the above parameters one compute the following metrics:

 Program Length: N = N1 + N2
 Program Vocabulary: n = n1 + n2
 Volume: V = N * lg n
 Difficulty: D = (n1 * N2) / (2 * n2)
 Effort: E = D * V
 Time to Implement: T = E / 18 (in seconds) [vi]
The program volume V is the minimum number of bits needed to encode the program.
It represents the size of the program while taking into account the programming
language.
The difficulty metric indicates how difficult a program is to write or understand.
Effort denotes the "mental effort" required to develop the software, or to recreate the
same in another programming language [iv].

Using Basic COCOMO model to estimate project parameters

Use the simulator on the right hand side to understand how project type and size affects
the different parameters estimated.
Quick glance at the formulae:
 Effort: a * (Size)b person-month
 Time for development: 2.5 * (Effort)c month
Drag the slider to change the project size. Note: select the nearest discrete value
corresponding to the actual size.
 Project Type a b c
2 1. 0.
Organic . 0 3
4 5 8
Project size (in KLOC) 2
Effort (in PM) 4.97
Tdev (in month) 4.6
# of developers 2

Project Type a b c
2 1. 0.
Organic . 0 3
4 5 8
Project size (in KLOC) 20
Effort (in PM) 55.76
Tdev (in month) 11.52
# of developers 5

Project Type a b c
2 1. 0.
Organic . 0 3
4 5 8
Project size (in KLOC) 40
Effort (in PM) 115.44
Tdev (in month) 15.19
# of developers
8

Project Type a b c
1. 0.
Organic 3 1 3
2 5
Project size (in KLOC) 200
1133.1
Effort (in PM)
2
Tdev (in month) 29.3
 Project Type a b c
# of developers 39

Project Type a b c
3 1 0.
Organic . . 3
6 2 2
Project size (in KLOC) 400
Effort (in PM) 4772.81
Tdev (in month) 37.6
# of developers 127

As evident from the simulation parameters, size of a semi-detached project is larger than
that of an organic project, and size of an embedded project is larger than that of a semi-
detached, and thereby affecting factors like effort and development time.

Case Study
[Hide]# 1 : A Library Information System for SE VLabs Institute
The SE VLabs Institute has been recently setup to provide state-of-the-art research
facilities in the field of Software Engineering. Apart from research scholars (students) and
professors, it also includes quite a large number of employees who work on different
projects undertaken by the institution.
As the size and capacity of the institute is increasing with the time, it has been proposed to
develop a Library Information System (LIS) for the benefit of students and employees of
the institute. LIS will enable the members to borrow a book (or return it) with ease while
sitting at his desk/chamber. The system also enables a member to extend the date of his
borrowing if no other booking for that particular book has been made. For the library staff,
this system aids them to easily handle day-to-day book transactions. The librarian, who has
administrative privileges and complete control over the system, can enter a new record
into the system when a new book has been purchased, or remove a record in case any book
is taken off the shelf. Any non-member is free to use this system to browse/search books
online. However, issuing or returning books is restricted to valid users (members) of LIS
only.
The final deliverable would a web application (using the recent HTML 5), which should run
only within the institute LAN. Although this reduces security risk of the software to a large
extent, care should be taken no confidential information (eg., passwords) is stored in plain
text.
The SE VLabs Institute has a IT management team of it's own. This team has been given the
task to execute the Library Information System project. The team consists of a few experts
from industry, and a batch of highly qualified engineers experienced with design and
implementation of information systems. It is planned that the current project will be
undertaken by a small team consisting of one expert and few engineers. Actual team
composition would be determined in a later stage.
Using COCOMO and based on the team size (small) and experience (high), the concerned
project could be categorized as "organic". The experts, based on their prior experience,
suggested that the project size could roughly be around 10 KLOC. This would serve as the
basis for estimation of different project parameters using basic COCOMO, as shown below:

Effort = a * (KLOC)b PM
Tdev = 2.5 * (Effort)c Months
For organic category of project the values of a, b, c are 2.4, 1.05, 0.38 respectively. So, the
projected effort required for this project becomes
Effort = 2.4 * (10)1.05 PM
= 27 PM (approx)
So, around 27 person-months are required to complete this project. With this calculated
value for effort we can also approximate the development time required:
Tdev = 2.5 * (27)0.38 Months
= 8.7 Months (approx)
So, the project is supposed to be complete by nine months. However, estimations using
basic COCOMO are largely idealistic. Let us refine them using intermediate COCOMO. Before
doing so we determine the Effort Adjustment Factor (EAF) by assigning approprite weight
to each of the following attributes.
Ratings

Ver Ver Extr

y y a
Lo Lo Nomin Hig Hig Hig
Cost Drivers w w al h h h

Product attributes

0.8
Required software reliability 0.75 1.00 1.15 1.40
8

0.9
Size of application database 1.00 1.08 1.16
4

0.8
Complexity of the product 0.70 1.00 1.15 1.30 1.65
5

Hardware attributes

Run-time performance constraints 1.00 1.11 1.30 1.66

Memory constraints 1.00 1.06 1.21 1.56

 Ratings

Ver Ver Extr

y y a
Lo Lo Nomin Hig Hig Hig
Cost Drivers w w al h h h

Volatility of the virtual machine environ 0.8

1.00 1.15 1.30
ment 7

0.8
Required turnabout time 1.00 1.07 1.15
7

Personnel attributes

1.1
Analyst capability 1.46 1.00 0.86 0.71
9

1.1
Applications experience 1.29 1.00 0.91 0.82
3

1.1
Software engineer capability 1.42 1.00 0.86 0.70
7

1.1
Virtual machine experience 1.21 1.00 0.90
0

1.0
Programming language experience 1.14 1.00 0.95
7

Project attributes

Application of software engineering me 1.1

1.24 1.00 0.91 0.82
thods 0

1.1
Use of software tools 1.24 1.00 0.91 0.83
0

1.0
Required development schedule 1.23 1.00 1.04 1.10
8

The cells with yellow backgrounds highlight our choice of weight for each of the cost
drivers. EAF is determined by multiplying all the chosen weights. So, we get
EAF = 0.53 (approx)
Using this EAF value we refine our estimates from basic COCOMO as shown below
Effort|corrected = Effort * EAF
= 27 * 0.53
= 15 PM (approx)
 Tdev|corrected = 2.5 * (Effort|corrected)c
= 2.5 * (15)0.38
= 7 months (approx)
After refining our estimates it seems that seven months would likely be sufficient for
completion of this project. This is still a rough estimate since we have not taken the
underlying components of the software into consideration. Complete COCOMO model
considers such parameters to give a more realistic estimate.

1. According to the COCOMO model, a project can be categorized into

3 types 5 types 5 types No such categorization
2. In Intermediate COCOMO model, Effort Adjustment Factor (EAF) is derived from the
effort multipliers by
Adding them Multiplying them Taking their weighted average
Considering their maximum
3. Project metrics are estimated during which phase?
Feasibility study Planning Design Development
4. According to Halsetad's metrics, program length is given by the
Sum of total number of operators and operands
Sum of number of unique operators and operands
Total number of operators Total number of operands
5. Complete COCOMO considers a software as a
Homogeneous system Heterogeneous system
6. Consider you are developing a web application, which would make use of a lot of web
services provided by Facebook, Google, Flickr. Would it be wise to make estimates for this
project using COCOMO?
Yes, of course Not at all

Steps for conducting the experiment

General Instructions
Follow are the steps to be followed in general to perform the experiments in Software
Engineering Virtual Lab.
1. Read the theory about the experiment
2. View the simulation provided for a chosen, related problem
3. Take the self evaluation to judge your understanding (optional, but recommended)
4. Solve the given list of exercises
Experiment Specific Instructions
Following are the instructions specifically for this experiment:
1. The left hand side of the 'Exercises' page will present the problem
2. The right hand side of the page asks to evaluate certain parameters
3. The values of the parameters are to be typed in in the adjoining text boxes
 4. If all the values entered are correct, then the solution is correct. Otherwise user will
be indicated where the error has occurred.

Considering your immense expertise in software development, The Absolute Beginners Inc.
has recently allotted you a mega project. The goal of the project is to create a database of all
Hindi films released since 2000. The software would allow one to generate a list of top ten
hit films, top ten flop films, best comedy films, and so on. Using your prior experience you
have decided the approximate sizes of each module of the software as follows:
 Data entry (0.9 KDSI) Data update (0.7 KDSI)
 Query (0.9 KDSI) Report generation and display (2 KDSI)
Also take into consideration the following cost drivers with their ratings:
 Storage constraints (Low)
 Experience in developing similar software (High)
 Programming capabilities of the developers (High)
 Application of software engineering methods (High)
 Use of software tools (High)
(All other cost drivers have nominal rating).
Now answer the following:
 Applying intermediate COCOMO estimate the minimum size of the team you would
require to develop this system
 Assuming that your client would pay Rs. 50,000 per month of development, how
much would be the likely billing?
Learning Objectives:
1. Identify type of a project as per COCOMO
2. Prepare an estimate of required effort and cost
Limitations: Values presented here are arbitrary and doesn't relate to real life

Note: The above example has been adapted from COCOMO (Constructive Cost Model),
Seminar on Software Cost Estimation WS 2002 / 2003, presented by Nancy Merlo – Schett.

Project Type a b c
2. 1.0 0.3
Organic 4 5 8

Project size (in KDSI)

Effort (in PM)
Tdev (in month)
Effort Adjustment Factor (EAF)
Effort|corrected (in PM)
Tdev|corrected (in month)
# of developers
Identify the unique operators and operands from the following snippet of code:
01 int

02 main(int argc, char **argv)

03 {

int
0
x=
4
10;

05 int y = 20;

06 int sum;

07

08 sum = x + y;

09

10 printf("Sum of %d and %d is: %d\n", x, y, sum);

1
1

1 retu
2 rn 0;

13 }

Note: In cases where you've to enter multiple values (for example, listing the operators
from a code snippet), please separate them with a comma

int main argc char * argv

() {}
Operators x = 10 ; y 20 sum +
printf "Sum of %d and %d is: %d " ,
return 0

Operands
int main argc char * argv
() {}
x = 10 ; y 20 sum +
 int main argc char * argv
() {}
Operators x = 10 ; y 20 sum +
printf "Sum of %d and %d is: %d " ,
return 0

printf "Sum of %d and %d is: %d " ,

return 0

The Absolute Beginners Inc. is again at your door! This time their demand is, however,
simple. They have a C program, which computes the area of a circle (code shown below).
They want it to be rewritten in Java.
0
int
1
main(int
0
argc, cha
2
r **argv)
03 {
04 int radius = 12.34;
05
06 printf("Area of the circle with radius %f is: %f\n", radius, area(radius));
0
7
0 retu
8 rn 0;
09 }
10
11 float
12 area(float r) {
13 return 22 * r * r / 7;
14 }
Using Halstead's metrics estimate the effort required to recreate this program.
Learning Objectives:
1. Determining estimated effort using Halstead's metrics
 Parameter Value
Total # of operators

Total # of operands

Total # of unique operators

Total # of unique operands

Program length
Program vocabulary
Volume
Difficulty
Effort
Time to implement (in seconds)

Following books and websites have been consulted for this experiment.
You are suggested to go through them for further details.
Bibliography
1. Fundamentals of Software Engineering, Rajib Mall, Prentice-Hall of India, 3rd
Edition, 2009
2. Software Engineering, Ian Sommerville, Addison Wesley Longman, 9th Edition,
March 2010
Webliography
i. COCOMO
ii. Halstead complexity measures
iii. COCOMO (Constructive Cost Model), Seminar on Software Cost Estimation WS
2002 / 2003, presented by Nancy Merlo – Schett
iv. The Halstead metrics
v. Software Engineering, National Program on Technology Enhanced Learning
vi. Halstead Metrics, Verifysoft Technology