Department of Computing

CRESTES

Continual Resource ESTimation for Evolving Software

CRESTES was a 15-month research project based at The Open University in the UK, and is funded by the Engineering and Physical Sciences Research Council (EPSRC) under grant no. GR/S90782.

Below you can find details of our goals and an overview of our research. By navigating through the pages of this website you can find out more about the people involved, our publications and activities that were undertaken during this short project. For any further information please contact either Juan Fernandez-Ramil or Sarah Beecham, or write to:
Department of Computing Faculty of Maths and Computing The Open University Walton Hall Milton Keynes MK7 6AA, UK
NEWS: Project official end: The CRESTES project ended in December 2005. The final report to the EPSRC was submitted in June 2006. Seeking continuation of the funding for further research into this topic: In the summer of 2006, we will re-submit a proposal to the EPSRC to extend this year project for three additional years. We are grateful to the EPSRC for giving CRESTES a non-cost extension till Christmas 2005 - this extra time allowed us to start work with some of our new collaborators.

THE CRESTES PROJECT: RESEARCH GOALS and OVERVIEW

The Continual Resource ESTimation for Evolving Software – CRESTES – project investigated a novel approach for resource and schedule estimation for long-lived software, a problem of key importance in a society ever more dependant on software. The approach is based on the observation that a software system evolves through distinguishable stages. The stages include periods of relatively stable functional growth and change rate, separated by transition periods of re-structuring. Our hypothesis was that each of the stable phases can be characterised by a distinct economic model. If this hypothesis holds for a variety of software systems and domains, the problem of deriving estimation models can be approached by (1) the identification of stable economic phases, (2) detailed economic modelling within each stage and (3) detection of the emergence of a transition period. The project has made progress towards an approach to resource estimation for the longest section of the software lifecycle, a portion for which there are no widely accepted approaches.

In future work we wish to refine our measurements and models based on exposure to further empirical data set and additional systematic experimentation with a variety of qualitative and quantitative data modelling techniques. In our interaction with our industrial collaborators we seek to develop an approach that is sufficiently flexible to address the resource estimation requirements of a wide variety of software evolution processes.

For further details contact: Juan Fernandez-Ramil or Sarah Beecham

Links

	Centre for Computing Research
	Computing Department
	Mathematics and Computing Faculty
	The Open University