Improving Software Effort Estimation Using an Expert-Centred Approach

  • Emilia Mendes
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7623)


A cornerstone of software project management is effort estimation, the process by which effort is forecasted and used as basis to predict costs and allocate resources effectively, so enabling projects to be delivered on time and within budget. Effort estimation is a very complex domain where the relationship between factors is non-deterministic and has an inherently uncertain nature, and where corresponding decisions and predictions require reasoning with uncertainty. Most studies in this field, however, have to date investigated ways to improve software effort estimation by proposing and comparing techniques to build effort prediction models where such models are built solely from data on past software projects - data-driven models. The drawback with such approach is threefold: first, it ignores the explicit inclusion of uncertainty, which is inherent to the effort estimation domain, into such models; second, it ignores the explicit representation of causal relationships between factors; third, it relies solely on the variables being part of the dataset used for model building, under the assumption that those variables represent the fundamental factors within the context of software effort prediction. Recently, as part of a New Zealand and later on Brazilian government-funded projects, we investigated the use of an expert-centred approach in combination with a technique that enables the explicit inclusion of uncertainty and causal relationships as means to improve software effort estimation. This paper will first provide an overview of the effort estimation process, followed by the discussion of how an expert-centred approach to improving such process can be advantageous to software companies. In addition, we also detail our experience building and validating six different expert-based effort estimation models for ICT companies in New Zealand and Brazil. Post-mortem interviews with the participating companies showed that they found the entire process extremely beneficial and worthwhile, and that all the models created remained in use by those companies. Finally, the methodology focus of this paper, which focuses on expert knowledge elicitation and participation, can be employed not only to improve a software effort estimation process, but also to improve other project management-related activities.


Software Effort Estimation Expert-centred Approach Process Improvement Cost Estimation Project Management 


  1. 1.
    Jørgensen, M., Grimstad, S.: Software Development Effort Estimation: Demystifying and Improving Expert Estimation. In: Tveito, A., Bruaset, A.M., Lysne, O. (eds.) Simula Research Laboratory - by thinking constantly about it, ch. 26, pp. 381–404. Springer, Heidelberg (2009)Google Scholar
  2. 2.
    Mendes, E.: Cost Estimation Techniques for Web Projects. IGI Global Publishers (2007)Google Scholar
  3. 3.
    Kitchenham, B.A., Pickard, L.M., Linkman, S., Jones, P.: Modelling Software Bidding Risks. IEEE Transactions on Software Engineering 29(6), 542–554 (2003)CrossRefGoogle Scholar
  4. 4.
    Jørgensen, M., Shepperd, M.J.: A Systematic Review of Software Development Cost Estimation Studies. IEEE Transactions Software Engeneering 33(1), 33–53 (2007)CrossRefGoogle Scholar
  5. 5.
    Azhar, D., Mendes, E., Riddle, P.: A Systematic Review of Web Resource Estimation. In: Proceedings of PROMISE 2012 (accepted for publication, 2012)Google Scholar
  6. 6.
    Mendes, E., Mosley, N., Counsell, S.: Investigating Web Size Metrics for Early Web Cost Estimation. Journal of Systems and Software 77(2), 157–172 (2005), doi:10.1016/j.jss.2004.08.034CrossRefGoogle Scholar
  7. 7.
    Ruhe, M., Jeffery, R., Wieczorek, I.: Cost estimation for Web applications. In: Proceedings ICSE 2003, pp. 285–294 (2003)Google Scholar
  8. 8.
    Ferrucci, F., Gravino, C., Di Martino, S.: A Case Study Using Web Objects and COSMIC for Effort Estimation of Web Applications. In: EUROMICRO-SEAA, pp. 441–448 (2008)Google Scholar
  9. 9.
    Mendes, E., Mosley, N., Counsell, S.: Web metrics - Metrics for estimating effort to design and author Web applications. IEEE MultiMedia, 50–57 (January-March 2001)Google Scholar
  10. 10.
    Reifer, D.J.: Web Development: Estimating Quick-to-Market Software. IEEE Software, 57–64 (November-December 2000)Google Scholar
  11. 11.
    Mendes, E.: Using Knowledge Elicitation to Improve Web Effort Estimation: Lessons from Six Industrial Case Studies. In: Proceedings of the International Conference on Software Engineering (ICSE 2012), track SE in Practice, pp. 1112–1121 (2012)Google Scholar
  12. 12.
    Mendes, E.: Knowledge Representation using Bayesian Networks A Case Study in Web Effort Estimation. In: Proceedings of the World Congress on information and Communication Technologies (WICT 2011), pp. 310–315 (2011)Google Scholar
  13. 13.
    Mendes, E.: Building a Web Effort Estimation Model through Knowledge Elicitation. In: Proceedings of the International Conference on Enterprise Information Systems (ICEIS), pp. 128–135 (2011)Google Scholar
  14. 14.
    Mendes, E., Polino, C., Mosley, N.: Building an Expert-based Web Effort Estimation Model using Bayesian Networks. In: 13th International Conference on Evaluation & Assessment in Software Engineering (2009)Google Scholar
  15. 15.
    Ammerman, M.: The Root Cause Analysis Handbook: A Simplified Approach to Identifying, Correcting, and Reporting Workplace Errors (1998)Google Scholar
  16. 16.
    Woodberry, O., Nicholson, A., Korb, K., Pollino, C.: Parameterising Bayesian Networks. In: Australian Conference on Artificial Intelligence, pp. 1101–1107 (2004)Google Scholar
  17. 17.
    Druzdzel, M.J., van der Gaag, L.C.: Building Probabilistic Networks: Where Do the Numbers Come From? IEEE Trans. on Knowledge and Data Engineering 12(4), 481–486 (2000)CrossRefGoogle Scholar
  18. 18.
    Tang, Z., McCabe, B.: Developing Complete Conditional Probability Tables from Fractional Data for Bayesian Belief Networks. Journal of Computing in Civil Engineering 21(4), 265–276 (2007)CrossRefGoogle Scholar
  19. 19.
    Jensen, F.V.: An introduction to Bayesian networks. UCL Press, London (1996)Google Scholar
  20. 20.
    Baker, S., Mendes, E.: Aggregating Expert-Driven Causal Maps for Web Effort Estimation. In: Kim, T.-H., Kim, H.-K., Khan, M.K., Kiumi, A., Fang, W.-C., Ślęzak, D. (eds.) ASEA 2010. CCIS, vol. 117, pp. 264–282. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  21. 21.
    Montironi, R., Whimster, W.F., Collan, Y., Hamilton, P.W., Thompson, D., Bartels, P.H.: How to develop and use a Bayesian Belief Network. Journal of Clinical Pathology 49, 194 (1996)CrossRefGoogle Scholar
  22. 22.
    Baker, S., Mendes, E.: Evaluating the Weighted Sum Algorithm for Estimating Conditional Probabilities in Bayesian Networks. In: Proceedings of the Software Engineering and Knowledge Engineering Conference (SEKE 2010), pp. 319–324 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Emilia Mendes
    • 1
  1. 1.School of ComputingBlekinge Institute of TechnologyKarlskronaSweden

Personalised recommendations