The results of the study have shown that over the last 20 years, the public sector has tended to be optimistic in its estimates for projects over £40m in value although there was evidence of improvement over the same time period. The degree of optimism was dependent on the type of project and the maturity of the business case.
Optimism developed as a result of failing to manage all project risks. The 'inadequacy of business case' was identified as the most critical project risk area, with risk arising from inadequate definition of project requirements and method of implementation, and inadequate attention to risk mitigation in developing the chosen option. There was also insufficient consideration of possible changes in the need for the project during the life of any assets or term of a contract.
Optimism bias for projects is not sector specific, as similar levels of optimism bias were recorded for project types across sectors. Some project types, where high levels of optimism bias were recorded, are inherently more risky than others. The following project types are listed in descending order of inherent risk, based on capital expenditure optimism bias:
1. Equipment/development
2. Non-standard civil engineering
3. Non-standard buildings
4. Outsourcing
5. Standard civil engineering
6. Standard buildings
There is no correlation between project size and optimism bias, however there is a strong relationship between project size and the number of project risks. Major projects like those in the Mott MacDonald study and minor projects (approximately £10 m in value) have the same number of project risk areas whose project risks need to be managed. The number of project risks within project risk areas increases with size of project. Optimism bias measures the level to which project risks are not managed (i.e. low optimism bias reflects a high percentage of managed project risks, while a high optimism bias represents a low percentage of managed project risks). Therefore the level of optimism bias recorded for a project will be dependant on the project management and risk management capabilities of the project management team rather that the number of risks associated with the project.
The management of project risks for major projects is likely to require more money and effort than that for smaller projects. However, since optimism bias is measured as a percentage increase of project outcomes compared with the business case estimates relevant to the appraisal, similar levels of optimism bias can be expected for major and minor projects.
The data collection exercise identified shortcomings in record keeping, post-completion benefit appraisal, and allocation of operating phase costs within most of the projects studied. Once a project was completed, archiving of its records tended to be disorganised and post project reviews were not performed. As a result, lessons learned on that project were lost.
"Those who do not learn from the past are condemned to repeat it" Anonymous
Therefore Mott MacDonald recommends that a process actively promoting knowledge transfer and knowledge sharing should be put in place. Adopting the following will allow continued improvements through the lessons learned from completed projects:
• An open approach to sharing the successes and failures of major project procurements, through internal and external seminars, papers and similar
• Post completion, one year after completion and five years after completion audits to compare project outturns against projections, together with wide dissemination of lessons learned
• Methodical archiving of key project documents.