Sassano, Giuseppe; (2022) Optimism bias and cost overruns: experimenting on the internal and external views in resources and time estimation. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text PhD Dissertation Giuseppe Sassano 15048738 removed signature.pdf - Other Download (3MB) | Preview

Abstract

Optimism bias affects most estimation based human decisions, from daily activities to the appraisal of big infrastructure projects. Building upon the underlying constructs of this behaviour through the lenses of support and prospect theories, operationalised in the internal and external view in the project management context, helped me to formulate a conceptual framework. This framework, called the “Holistic view” suggests ways to integrate the two perspectives with the aim to improve the quality of forecasts in infrastructure management. Integration can be achieved by including subjective probabilities and unpacking techniques into case-based reasoning methods. Based on this framework and on the analysis of the relevant policies, perspectives, definitions and techniques provided by the literature on optimism bias in project forecasting, I designed and administered four experiments, with a sample of 231 participants. The results of the first experiment show that there is a positive relationship between different levels of dispositional optimism and resource overruns. In the second experiment, I found that adding an optimism uplift to an estimate structurally increases forecast precision, however, this might lead to the use of more resources than when optimism uplift is not applied. In the third experiment, I found that unpacking, whilst making estimations slightly more precise, does not influence the estimations as in the previous case, but only the estimator. The results of the fourth experiment, combining unpacking with optimism uplift, indicated that forecast precision starkly increased, supporting the adoption of an “Holistic view”. This research shows the relevance of experimental methods in project management, unveils new relationships between different perspectives in the project forecasting field, and analyses, net of other factors, the impact of forecasting tools integrated in some of the policies addressing the effect of optimism bias in estimations, suggesting ways to improve those and the overall forecasting process of infrastructure projects.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item