ESTIMATING STEPWISE AND FUZZY REGRESSION ANALYSIS FOR MODELLING SOFTWARE DESIGN PROJECT RISKS
ABDELRAFE ELZAMLY *
Department of Industrial Computing, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka (UTeM), Malaysia and Department of Computer Science, Faculty of Applied Sciences, Al-Aqsa University, Palestine
Department of Industrial Computing, Faculty of Information and Communication Technology, Universiti Teknikal Malaysia Melaka (UTeM), Malaysia
*Author to whom correspondence should be addressed.
Software projects are normally associated with failure risk. This risk is not avoidable in a software design project, but it can be mitigated and controlled with various approaches. Taking into consideration, the software managers and IT managers depend on their experience and judgment to mitigate software project design risks in the previous studies. However, the software risk factors usually a common in a software design project and the experience among project managers is also varies, thus one need to think other approach to mitigate software design risks. This study aimed to test the level of prediction between two model scenarios namely- stepwise and fuzzy multiple regression analysis by using MMRE and Pred (l). The results show that, there is a significant relationship between them and possibly can be used to mitigate software project design risks. All models in software design process were acceptable value for MMRE and Pred (0.25) except model 3.
Keywords: Software development project risk management, design process, mean magnitude of relative error, prediction at Level L