Effort Drivers in Maintenance Outsourcing - An Experiment Using Taguchi's Methodology

Authors:
Baru S. Rao;N. L. Sarda
Affiliations:
-;-
Venue:
CSMR '03 Proceedings of the Seventh European Conference on Software Maintenance and Reengineering
Year:
2003

Citing 0
Cited 3

Estimating software maintenance effort: a neural network approach

ISEC '08 Proceedings of the 1st India software engineering conference
Multi-variate principal component analysis of software maintenance effort drivers

ICCSA'10 Proceedings of the 2010 international conference on Computational Science and Its Applications - Volume Part II
Dynamic software maintenance effort estimation modeling using neural network, rule engine and multi-regression approach

ICCSA'12 Proceedings of the 12th international conference on Computational Science and Its Applications - Volume Part IV

Quantified Score

Hi-index	0.00

Visualization

Abstract

We establish the context of software maintenanceoutsourcing of large commercial systems and discuss theproblem of estimating the maintenance effort in absenceof key parameters such as development effort andmaintenance history. To arrive at the drivers that couldhave a significant effect on the maintenance effort, anexperiment was conducted at Syntel, a NASDAQ listedapplication management and e-business solutionscompany. We present the list of possible drivers, theprofile of the people who participated in the experiment, the details of the 36 experiments and the statisticalanalysis of the experiment based on Taguchi'smethodology. For a system whose size parameter interms of lines of code is known, we conclude that the fivemost significant parameters that effect the efforts ofmaintenance are a)the multi time zone support, b)average number of lines per program, c)percentage ofonline programs in the total system, d)the nature ofservice level agreements and e)the complexity of the filesystem being used.