EKEdouard Kujawski

Edouard Kujawski

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Edouard Kujawski has a total of 2 co-authors. Here are the top co-authors by number of shared publications:

  • Mariana L. Alvaro

    Department of Statistics, California State University, Hayward, CA 94542

    1 shared publication
  • William R. Edwards

    Engineering Division, Ernest Orlando Lawrence Berkeley National Laboratory, University of California, Berkeley, CA 94720

    1 shared publication

Latest Publications

  • Incorporating psychological influences in probabilistic cost analysis

    by Edouard Kujawski, Mariana L. Alvaro, William R. Edwards

    Today's typical probabilistic cost analysis assumes an “ideal” project that is devoid of the human and organizational considerations that heavily influence the success and cost of real‐world projects. In the real world “Money Allocated Is Money Spent” (MAIMS principle); cost underruns are rarely available to protect against cost overruns while task overruns are passed on to the total project cost. Realistic cost estimates therefore require a modified probabilistic cost analysis that simultaneously models the cost management strategy including budget allocation. Psychological influences such as overconfidence in assessing uncertainties, dependencies among cost elements, and risk are other important considerations that are generally not addressed. It should then be no surprise that actual project costs often exceed the initial estimates and are delivered late and/or with a reduced scope. This paper presents a practical probabilistic cost analysis model that incorporates recent findings in human behavior and judgment under uncertainty, dependencies among cost elements, the MAIMS principle, and project management practices. Uncertain cost elements are elicited from experts using the direct fractile assessment method and fitted with three‐parameter Weibull distributions. The full correlation matrix is specified in terms of two parameters that characterize correlations among cost elements in the same and in different subsystems. The analysis is readily implemented using standard Monte Carlo simulation tools such as @Risk and Crystal Ball ® . The analysis of a representative design and engineering project substantiates that today's typical probabilistic cost analysis is likely to severely underestimate project cost for probability of success values of importance to contractors and procuring activities. The proposed approach provides a framework for developing a viable cost management strategy for allocating baseline budgets and contingencies. Given the scope and magnitude of the cost‐overrun problem, the benefits are likely to be significant. © 2004 Wiley Periodicals, Inc. Syst Eng 7: 000–000, 2004


  • Selection of technical risk responses for efficient contingencies

    by Edouard Kujawski

    The primary goal of good project risk management should be to successfully deliver projects for the lowest cost at an acceptable level of risk. This requires the systematic development and implementation of a set of Risk Response Actions (RRAs) that achieves the lowest total project cost for a given probability of success while meeting technical performance and schedule. We refer to this set as the “efficient RRA set.” This work presents a practical and mathematically sound approach for determining the efficient RRA set. It builds on some of Markowitz's portfolio selection principles and introduces several conceptual and modeling differences to properly treat project technical risks. The set of RRAs is treated as whole and not just individual risks. The efficient RRA set is determined based on “Outcome Cost vs. Probability of Success.” The risks and RRAs are characterized using scenarios, decision trees, and cumulative probability distributions. The analysis provides information that enables decision‐makers to select the efficient RRA set that explicitly takes their attitude toward project risk into account. Decision‐makers should find it both useful and practical for sound decision‐making under uncertainty/risk and efficiently optimizing project success. The computations are readily performed using commercially available Monte Carlo simulation tools. The approach is detailed using a realistic but simplified case of a project with two technical risks. © 2002 Wiley Periodicals, Inc. Syst Eng 5, 194–212, 2002


  • Why projects often fail, even with high cost‐contingencies

    by Edouard Kujawski

    In this note we assume that the individual risks have been adequately quantified and the total project cost contingency adequately computed to ensure an agreed‐to probability or confidence level that the total project cost estimate will not be exceeded. But even projects that implement such a process are likely to result in significant cost overruns and/or project failure if the project manager allocates the contingencies to the individual subsystems. The intuitive and mathematically valid solution is to maintain a project‐wide contingency and to distribute it to the individual risks on an as‐needed basis. Such an approach ensures cost‐efficient risk management, and projects that implement it are more likely to succeed and to cost less. We illustrate these ideas using a simplified project with two independent risks. The formulation can readily be extended to multiple risks. © 2002 Wiley Periodicals, Inc. Syst Eng 5: 151–155, 2002