A study on the distribution and cost prediction of requirements changes in the software life-cycle

  • Authors:

  • Chengying Mao;Yansheng Lu;Xi Wang

  • Affiliations:

  • College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, P.R. China;College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, P.R. China;College of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, P.R. China

  • Venue:
  • SPW'05 Proceedings of the 2005 international conference on Unifying the Software Process Spectrum
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

Software development is a dynamic process. Requirements change (RC) is inevitable and brings great challenges to the software development. How to precisely predict requirements change is especially important in the field of requirements engineering. In this paper, an assessment framework for the factors of RCs’ distribution is constructed firstly. Apart from the rough prediction method based on the statistic process control of RCs, an artificial neural network method for predicting RCs’ distribution is presented. In this case, the weight of each factor is calculated by a fuzzy logic method, called experts ranking. Furthermore, we propose a model to pre-evaluate the cost caused by RCs. With some practical projects data, a validation experiment has been drawn, whose result shows that our method and model are practical and efficient to predict the distribution and cost of RCs.