How software size influence productivity and project duration

Abstract

To make a perfect project plan, the software size of the order from the customer is the most important factor. The biggest challenge for the project manager is to estimate the project end date in the beginning of the project i.e. in project planning phase with realistic accuracy. Apart from other major inputs to estimate the project end date, expected team capability (productivity) and estimated software size are the major inputs that may influence the project end date. Software size is one of the most significant independent metric available in the planning phase and project manager has to estimate the other metrics based on the initial estimated software size. There is no direct relationship available between software size and project duration or software size and team productivity, however, there are industry data published by Quantitative Software Management and ISBSG that shows how these metrics influence each other. In this paper, using the data published by ISBSG and Quantitative Software Management, we try to statistically establish how productivity and project duration are influenced by software size. We have done linear regression analysis by generating the secondary data based on the data published by ISBSG and Quantitative Software Management. Linear regression equation validated with the actual project data and experimental results suggest that that productivity is significantly dependent on software size, however, project duration does not significantly depend on software size but may also be dependent on other metrics like team size, apart from software size.