Thanks:  0
Likes:  0
Dislikes:  0

1. ## Estimations Tools

Hi,

Is the first time that I post here. I´m from brasil so my i can have some mistakes in my post.

I need some estimations tools, based in function points. Can someone help me?

Tks

2. ## Re: Estimations Tools

SEER-SEM 7.0 from a company called Galorath

www.galorath.com

3. ## Re: Estimations Tools

Testing Effort Estimation:

1. System study
2. scope, approach and estimations ( Proposal )
3. Test Plan Design
4. Test Case Design
5. Test Case Review
6. Test Case Execution
7. Defect Handling
8. Deliverables

1. System Study, for this task, time given to understand the application is dependent on the client company. This will be ranging from 0- 20 working days.
2. Proposal, Using the data collected while system study, company will send a proposal. In this proposal they will mention the time taken for completing the 3-8 tasks.
3. Test Plan Design, on average user will get 5-7 working days.
4. Test Case Design, here the actual calculations starts…

Test Case Design depends on the following parameters
1. LOC, lines of code
2. FP, Functional Points

100 LOC = 10 FP

for each FP, user can write minimum 3 test cases using the following techniques

1. Equivalence class
2. Boundary value analysis
3. error guessing

so, if there are 200 fp, user can write (200 * 3) = 600 test cases

** User can write 30 TC,s effectively / day.

Therefore no.of days to write TC’s = 600/30 = 20 days.

5. Test case Review, formula to find this task is:
TCR effort = ½ ( TC design )
6. Test Case execution:
TCE = 3/2 ( TC design )

7. Defect Handling::

Defect handling consists of
D submitting
D Assigning
D fixing/rejected/postponing
D closing

· 30 defects can be detected for 100 functional points / 1000 LOC , in the first build.

Simple formula: Dh effort = TC design effort

8. Deliverables, it consists of

Test Plan
Test Scenarios
Defect Reports
S/w Req vs TC design
TC design vs Defects reported
Defects Reported Vs Closed
Stability charts

Effort = 1/3 ( TC design)

Finally, effort can be calculate by adding the effort of all the above 8 steps

Note: till now , we have not considered any RISK in the effort.

Kind of Risks:

1. Unavailability of the application
2. Incompleteness of application
3. Unstable Integrity
4. Data Unavailability
5. Network down …. List continuous

So, we need to take the buffer for all this:

Buffer days = ¼ ( Total effort)

All this can be presented in the proposal.

4. ## Re: Estimations Tools

While this is a reasonable method assuming coding has taken place, I have been involved in new projects where I have designed test cases straight from requirements, because the project had not even progressed to the coding stage.

How would you estimate in the event that no coding had taken place.

5. ## Re: Estimations Tools

Strangely enough I have been hunting around the web for similar, and have stumbled on TPA (Test Point Analysis)

http://www.sogeti.nl/index.html?/iospagina.cfm?uNr=584
http://www.sogeti.nl/index.html?/iospagina.cfm?uNr=566
http://www.bfpug.org/fpug_rio/PCSQ/dekkers.pdf

6. ## Re: Estimations Tools

Originally posted by thatavarthi:
Testing Effort Estimation:
8. Deliverables, it consists of

Test Plan
Test Scenarios
Defect Reports
S/w Req vs TC design
TC design vs Defects reported
Defects Reported Vs Closed
Stability charts

All this can be presented in the proposal.
<font size="2" face="Verdana, Arial, Helvetica">HI,thatavarthi,
could you explain the "S/w Req" ? I don't understand what it's meaning. thx!

[ 02-03-2004, 09:48 PM: Message edited by: wasonzh ]

7. ## Re: Estimations Tools

Please see my caveats about the difficulties and misuse of function points -- and test points too -- which I just posted to another thread in this forum.

I advise a healthy skepticism about the estimating tools. You can obtain virtually any answer you want if you know how to use them.

Always cross-check the tool's estimate by independently calculating an estimate by another method. (I have an inventory of approximately 30 test estimating methods, not including the software estimating tools.)

Ross

8. ## Re: Estimations Tools

Hi Thatavarthi

What evidence do you have that the impressive array of numbers you present are reasonably correct? I suspect that they are nonsense, or dependent on so many hidden context-dependent variable factors that they are dangerous. I hope to be proven wrong.

Ross

9. ## Re: Estimations Tools

Thatavarthi,

I have to question any estimation technique that says test case execution takes 3/2 as long as test case design. It generally takes twice as long to document test cases as it does to perform them; sometimes even less...

Could you explain? Maybe there's something I don't understand.

Also, the old LOC/function point methodology is more difficult to defend with web technology; has anyone else found that "lines of code" doesn't have as much significance outside of the mainframe environment?

- Linda

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•
Search Engine Optimisation provided by DragonByte SEO v2.0.36 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.