Expert Judgment in Software Estimation During the Bid Phase of a Project -- An Exploratory Survey
Authors:
- Pedro FariaPFPedro Faria
- Eduardo MirandaEMEduardo Miranda
- 1 October 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 126-131
Abstract
This article presents the results of an industrial study about the reliability of expert judgment in cost estimation in a medium-sized software company. The purpose of the study was to assess current practices within the company, and to use the results as a catalyst for improving the company's cost estimation processes. The study included the analysis of two aspects: variability and calibration. First, the degree of variability can indicate the absence of a sound and repeatable process for making estimations. Second, measuring the estimators' calibration allowed us to establish their awareness about their lack of specific knowledge when estimating, as well as their ability to properly acknowledge and represent uncertainty. The results clearly show that, in the context of the organization studied, cost estimation based on unstructured expert judgment is unreliable.Keywords
This publication has 11 references indexed in Scilit:
- How to Measure Anything2010
- Protecting Software Development Projects against UnderestimationProject Management Journal, 2008
- A study of expert overconfidenceReliability Engineering & System Safety, 2008
- From Natural Language to Formal Proof GoalLecture Notes in Computer Science, 2006
- Better sure than safe? Over-confidence in judgement based software development effort prediction intervalsJournal of Systems and Software, 2004
- Overconfidence in Judgmental Forecasting2001
- Improving subjective estimates using paired comparisonsIEEE Software, 2001
- Building intelligent systems one e-citizen at a timeIEEE Intelligent Systems and their Applications, 1999
- Physicians' use of probabilistic information in a real clinical setting.Journal of Experimental Psychology: Human Perception and Performance, 1981
- Reliability of Subjective Probability Forecasts of Precipitation and TemperatureJournal of the Royal Statistical Society Series C: Applied Statistics, 1977