MELJUN CORTES JEDI Slides-7.6 Software Metrics and Estimation

7/29/2019 MELJUN CORTES JEDI Slides-7.6 Software Metrics and Estimation

1/22

Software Engineering 1

Basic Software Project

Management

Software Metricsand Estimation

TOPIC SIX


2/22


Software Metrics

It refers to a variety of measure used in computer software..

Measurements can be:

applied to software process with the intent of continuous

improvement. used throughout a software project to assist in estimation, quality

control, productivity assessment and project control.

used by software engineers to help assess the quality of technicalwork products and to assists in the tactical decision making as

project proceeds.


3/22


Software Measurement

Reasons to indicate the quality of the product

to assess the productivity of the people who produce theproduct

to assess the benefits derived from new softwareengineering methods and tools

to form a baseline for estimation

to help justify request for new tools or additional training.


4/22


Categories of Measurements

Direct Measurements

Software process measurements are cost and effort applied to thedevelopment of the software.

Product measurements are lines of code (LOC) produced,execution, speed, memory size and defects report over a period oftime.

Indirect Measurements

Product measurements are functionality, quality, complexity,efficiency, reliability, maintainability etc.


5/22


Metrics Categorization

Size-oriented Metrics

It is used to collect direct measures of software outputand quality based on the line of codes (LOC) produced.

Function-oriented Metrics It is used to collect direct measures of software output

and quality based on the program's functionality orutitlity.

Human-oriented Metrics It provides measures collected about the manner in

which people develop computer software and humanperception about the effectiveness of methods and tools.


6/22


Metrics Categorization Productivity Metrics

It refers to the measure on the output of softwareprocess.

Quality Metrics

It refers to the measure that provides indication of howclosely software conforms to implicit and explicitcustomer requirements.

It is used for software fitness.

Technical Metrics

It refers to the measure on the characteristic of theproduct such as logical complexity and degree ofmodularity.


7/22


Size-oriented Metrics (LOC)

It consists of a table which keeps track of the followinginformation on each software project done by anorganization:

effort in terms of man-days or person-month

cost

lines of code generated

pages of documentation

number of people


8/22


Size-oriented Metrics (LOC)

Project Effort Cost KLOC Pages Errors People

(in P)

Project One 22 222 .222 222 2 2 2

Project Two 22 222 .2 2 2 2222 22 2

Project Three 22 222 .222 2222 22 2

Project Four 22 22 22 22 22 2

Derived Metrics:Productivity = KLOC / person-months

Quality = errors / KLOC

Cost = Cost / KLOC


9/22


Function-oriented Metrics:

Function Points (FP) focus on the program's functionality or utility

are derived using an empirical relationship based oncountable measures of software's information domain and

assessment of software complexity used for business information system applications


10/22


Computation of Function

Point Determine the value of the information domain. Determine the complexity adjustment value.

Use the Function Point formula.


11/22


Information Domain

Weighing Factor

Measurement Count Simple Average Complex

Number of User Input x 2 2 2 =

Number of User Output x 2 2 2 =

Number of User Inquiries x 2 2 2 =

Number of Files x 2 22 22 =

Number of External Interfaces x 2 2 22 =

Count-Total =


12/22


Complexity Adjustment Value2 2 2 2 2 2

No Influence Incidental Moderate Average Significant Essential

Fx:1. Does the system requirereliable backup?

2. Are data communications required?3. Are there distributed processing functions?4. Is performance critical?5. Will the system run in an existing, heavily utilized operational environment?6. Does the system require on-line data entry?7. Does the on-line data entry require the input transaction to be built over multiplescreens or operations?8. Are the master files updated on-line?

9. Are the inputs, outputs, files, or inquiries complex?10. Is the internal processing complex?11. Is the code design to be reusable?12. Are conversion and installation included in the design?13. Is the system designed for multiple installations in different organizations?14. Is the application designed to facilitate change and ease of use by the user?


13/22


Function Point Formula

FP = Count-Total x [0.65 + 0.01 x sum(Fx)]


14/22


Function Points

Derived Metrics:

Productivity = FP / person-months

Quality = Defects / FP

Cost = Cost / FP


15/22


Reconciling LOC and FP

Program m ing LanguagesLOC/FP (Average)

Assem bly Language 222

C 222

Cobol 22

Ada 2 2 2

Java 22

Perl 2 2

Visual Bas ic 22

SQL 22C++ 22


16/22


Software Estimation

Requirements

experience

access to good historical information

courage to commit to quantitative measures when qualitative dataare all that exists


17/22


Using FP and LOC

Assume that the project similar to PROJECT FOUR of theSoftware Metrics History.

Steps:

Compute the Function Point. Determine the estimated LOC.

Use Derived Formulas for the effort and cost estimation.


18/22


Compute the Function Point

Information Domain Weighting FactorMeasurement Parameter Count S imple Average Complex

Number of User Inputs 2 x 3 4 6 6

Number of User Outputs 5 x 4 5 7 20Number of User Inquiries 3 x 3 4 6 9

Number of Files 2 x 7 10 15 14

Number of External

Interfaces

2 x 5 7 10 10

Count-Total 59


19/22


Compute the Function Point

2. Are data communications required? 3

3. Are there distributed processing functions? 1

4. Is performance critical? 2

5. Will the system run in an existing, heavily utilized operational environment? 3

6. Does the system require on-line data entry? 4

7. Does the on-line data entry require the input transaction to be built over multiplescreens or operations? 2

8. Are the master files updated on-line? 2

9. Are the inputs, outputs, files, or inquiries complex? 1

10.Is the internal processing complex? 1

11.Is the code designed to be reusable? 4

12.Are conversion and installation included in the design? 3

13.Is the system designed for multiple installations in different organizations? 1

14.Is the application designed to facilitate change and ease of use by the user? 1

Total= 31

FP = 59 X [0.65 + 0.01 x 31] = 56.64 FP


20/22


Determine the LOC

The project heavily uses JAVA.

LOC = 56.64 * 63 = 3568.32 LOC


21/22


Derive effort and cost

estimatesCost Estimates are computed as follows:

Cost of Producing a single LOC (From the Database): KLOC / Cost

Cost/KLOC = P30,000/15,000LOC = P2.00/LOC

Cost of the project = 3568.32 * 2.00 = P7136.64

Effort Estimates are computed as follows:

Number of LOC produced in a month (From the Database): KLOC/EffortKLOC/Effort = 15,000LOC / 8 months = 1875 LOC / month

Effort of the project = 3568.32 / 1875 = 1.9 or 2 months


22/22


Summary

Software Metrics

Categories of Measurement

Size-oriented Metrics

Lines Of Code (LOC)

Software Metric History

Function-oriented Metrics

Function Points (FP

Reconciling LOC and FP Metrics Project Estimations

MELJUN CORTES JEDI Slides-7.6 Software Metrics and Estimation

Documents

Transcript of MELJUN CORTES JEDI Slides-7.6 Software Metrics and Estimation