Dear Kiran, I use COnstructive COst MOdel:

http://csse.usc.edu/csse/research/COCOMOII/cocomo_main.html

https://files.ifi.uzh.ch/rerg/arvo/courses/seminar_ws02/reports/Seminar_4.pdf

I hope you find it useful.

Regards,

Kiran,

you can read the following articles about software cost estimation models:

[1] Lionel C. Briand, Khaled El Emam, Dagmar Surmann, Isabella Wieczorek, and Katrina D. Maxwell. 1999. An assessment and comparison of common software cost estimation modeling techniques. In Proceedings of the 21st international conference on Software engineering (ICSE '99), 313-322. DOI=10.1145/302405.302647.

[2] Chris F Kemerer. 1987. An empirical validation of software cost estimation models.Communications of the ACM 30, 5 (May 1987), 416-429. DOI=10.1145/22899.22906.

[3] Barbara A. Kitchenham , Emilia Mendes , Guilherme H. Travassos. 2007. Cross versus Within-Company Cost Estimation Studies: A Systematic Review, IEEE Transactions on Software Engineering 33, 5, 316-329.

[4] Tirimula Rao Benala, Satchidananda Dehuri, and Rajib Mall. 2012. Computational intelligence in software cost estimation: an emerging paradigm. ACM SIGSOFT Software Engineering Notes 37, 3 (May 2012), 1-7. DOI=10.1145/180921.2180932.

[5] Kevin Strike, Khaled El Emam, and Nazim Madhavji. 2001. Software Cost Estimation with Incomplete Data. IEEE Transactions on Software Engineering 27, 10 (October 2001), 890-908. DOI=10.1109/32.962560

Regards,

Zeljko

Thank you so much Mr. Edgar Serna M and Mr. Zeljko Stojanov 

 The people who can help you are your development team. Ask them. In my experience, most of the methods suggested above work only in commodity staff settings or are self-fulfilling prophecies.

Thank you so much Mr. James O Coplien

Dear Kiran,

I agree with James your best starting point is to do with locally avaialble and relevant data.

So I'd only use a model such as COCOMO if you really believe it captures aspects that are important.  It also suffers from the fact that the main size driver (KDSI) has to be estimated which may not be that easy. 

Other approaches use function points or use cases.

However whatever you do start simple and try to use as much historical data as possible.

Good luck! Martin

It depends on whether you plan to use LOCs or Functionalities as input for the estimation process. If you plan to use software functionalities as starting point, you can use some FP variant as a proxy for doing COCOMO cost estimation.