Predictive modeling of effort deviation for software projects  

Oleh:

Brar, Gurdarshan Singh ; Dubey, Pratibha Dikshit ; Ramasubrahmanyan, Sugathan

Jenis: Article from Proceeding
Dalam koleksi: ANQ Congress 19-22 October 2010 New Delhi, page 1-9.
Topik: IT Projects Effort Deviation; Project Outcome; Logistic Regression; Neural Network; Support Vector Machines
Fulltext: GS_Brar_Wipro_Predictive_Modeling.pdf (107.85KB)

Isi artikel

The Software Industry is one of the leading industries in India which is growing at a rapid pace. High Attrition, Changing Technologies, Peer Competition, Competition from other Low Cost Centers and the like are the centrifugal forces acting on this fast paced, people dependent Industry. As the saying goes “Time is Money”. It is of highest importance that the projects are completed On Time with no Effort Deviation or Cost Impact (Cost being directly proportional to the Effort Expended). Many parameters can impact Project Cost (tracked via the Metric - Effort Deviation) viz., Team Composition, Technology, Tools Usage, Code Reuse, Project Size, choice of Project Lifecycle, Effort Distribution across the phases of the project lifecycle, etc. A greater understanding of these parameters, their interaction, tweaking them for better predictability of the project outcome has become a necessary skill for a Project Manager or Program Manager and the organization at large. We at Wipro have developed a working model which helps in prediction of Project Cost using Neural Networks based on Effort Deviation. This paper details the factors used in modeling aspects, output and disseminating the information in the respective SBUs to assist project managers and corresponding Business Partners.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali