曝光台 注意防骗
网曝天猫店富美金盛家居专营店坑蒙拐骗欺诈消费者
4 Nadim F. Matta and Ronald N. Ashkena, ‘Why Good Projects Fail Anyway’, Harvard Business Review On
Line, (September 2003).
5 See James Reason, Managing the Risk of Organisational Accidents, (Ashgate, Aldershot, 1997).
6 A recent conference of the Major Projects Association highlighted poor risk management as the culprit. See
http://www.majorprojects.org/pubdoc/739.pdf .
Why Projects Fail
5
involvement in projects as consultants, the authors have come to the view that programme failures are
chiefly the result of technological factors, exacerbated by some cultural issues. The sheer scale of
failure suggests that the problems are endemic. The vast majority of the individuals managing projects
are well qualified, well trained and highly committed. Therefore the percentage of programme failures
clearly suggests a cause outside the sphere of human competence. Something basic and fundamental
is wrong.
Outmoded Technology
This paper advances the proposition that the most significant weakness in large engineering projects
today relates to the programme management and planning tools that are being deployed. In essence
the underlying methodology and philosophy of these tools has not changed a great deal in 50 years.
Most of them, in fact, were developed by the US military in the 1950s, as part of a wider adoption of
quantitative risk management techniques for defence procurement projects.7 According to Peter
Morris, one of the UK’s leading experts, project management is, ‘a discipline that remains stuck in a
1960s time warp’.8 As we will demonstrate the evolving characteristics of current programmes have
stretched the capabilities of current tools to breaking point. In our observations of the aerospace
industry this became crystal clear when something major went wrong, as the crisis management that
ensued invariably revealed very little detailed knowledge of the inner logic and structure of the project.
The key finding that we wish to advance here is that the projects have become opaque to those
managing them.
It would have been useful to the argument at this juncture to review a critical literature on the
weaknesses of quantitative project management techniques. Yet, as the author of a recent RAND
Corporation study discovered, no such critical literature exists.9 This is probably for two reasons. First,
the interested parties in project management are highly fragmented, being split across a practitioner
community and a variety of academic disciplines that publish in very disparate media. Secondly, the
failures of projects are highly sensitive and often involve proprietary information, which the analyst
cannot access. As a result we must confine ourselves here to some generic points and to observations
derived form our own participatory experience.
A highly significant issue that we encountered in our research was that there was little consensus
amongst practitioners about how best to manage large projects. In projects where quantitative risk
analysis was undertaken there seemed to be a general acceptance of some form of the critical path
method (CPM), but many of our interviewees were aware of the weaknesses of the deterministic and
linear modelling of CPM. 10 CPM tries to model the relationship between the linked tasks that
determine the overall length of the project. But the network generated by CPM is not a realistic
representation of task dependencies in large projects. When we asserted that modern engineering
programmes have become too large and too complex for the current generation of project tools to
adequately model we found few dissenters in our community of interviewees. Although, interestingly,
we did find a reluctance to embrace new techniques.
Representation and complexity
We return to this theme many times, but the underlying problem of project management we have
identified is an inability to accurately model the real tasks that comprise project work and the
dependencies between them. Current generation project management tools cannot represent the new
reality of large, complex and globally distributed projects. But in the design phase of projects the
process of iteration has never been adequately captured by tools that plot a linear critical path along a
7 See, J. Galway, ‘Quantative Risk Analysis for Project Management’, (RAND Corporation, Santa Monica,
2004).
8 Peter Morris, ‘Rethinking Project Management’, EPSRC Network 2004-2006 (UCL, 15 July 2004).
9 J. Galway, ‘Quantative Risk Analysis for Project Management’, (RAND Corporation, Santa Monica, 2004).
10 This is discussed in Galway, ibid.
Why Projects Fail
6
line of dependencies.11 Information/action dependencies are very complex and characterised by a
中国航空网 www.aero.cn
航空翻译 www.aviation.cn
本文链接地址：
航空资料41(74)
