| Software cost estimation |
|
|
 Construction Cost Model method comes from the spiral model for project planning, which defines four quadrants in each turn only one to three for the development and project management. The Construction Cost Model first in 1981, and was developed by Dr. Barry Boehm to estimate the cost, in number of man-months, and time to develop a software product. Originally it was built on a study of 63 software projects from 2000 to 100,000 lines of code in the company TRW Inc... But only one company has been fairly representative as a basis for development of Construction Cost Model? In addition, it is much related to the number of lines of code, especially the base model, but most programmers are experts (and their high salary); they write fewer lines of code for the same project! Construction Cost Model to the advantage of being open. Calibration data, formulas and details of the definitions are available. Participation in its development is encouraged. Today, Construction Cost Model II is a new product more suited to the aspect component reuse (existing modules). The 1998 version was calibrated on 161 data points, using the statistical approach 'Bayesian' to combine empirical data with expert advice on software cost estimation . In addition it can be re-calibrated on data from the company. A new model called Construction Cost Model is being developed by the team of Construction Cost Model. This model estimates the costs and planning of software projects using existing components. It is still a research project. Benefits: Construction Cost Model is the benchmark for detailed cost estimates and especially the breakdown of costs according to the phases of projects. Construction Cost Model estimates is more reliable, that the project settings are well known, that is to say we took advantage of previous projects to calibrate these parameters. |
| < Prev | Next > |
|---|
