The objectives of this thesis are to develop an analytical method for economic risk quantification during feasibility analysis for large engineering projects and to computerize
the method to explore its behavior, to validate it and to test its practicality for the measurement of uncertainty of decision variables such as project duration, cost, revenue, net present value and internal rate of return. Based on the probability of project success the method can be utilized to assist on strategic feasibility analysis issues such as contingency provision, "go-no go" decisions and adopting phased or fast track construction.
The method is developed by applying a risk measurement framework to the project economic structure. The risk measurement framework is developed for any function Y = g(X), between a derived variable and its correlated primary variables. Using a variable transformation, it transforms the correlated primary variables and the function
to the uncorrelated space. Then utilizing the truncated Taylor series expansion of the transformed function and the first four moments of the transformed uncorrelated
variables it approximates the first four moments of the derived variable. Using these first four moments and the Pearson family of distributions the uncertainty of the derived variable is quantified as a cumulative distribution function. The first four moments for the primary variables are evaluated from the Pearson family of distributions
using accurate, calibrated and coherent subjective percentile estimates elicited from experts. The correlations between the primary variables are elicited as positive definite correlation matrices. The project economic structure describes an engineering
project in three hierarchical levels, namely, work package/revenue stream, project
performance and project decision. Each of these levels can be described by Y = g(X), with the derived variables of the lower levels as the primary variables for the upper level. Therefore, the input as expert judgements is only at the work package/revenue stream level.
Project duration is estimated by combining the generalized PNET algorithm to the project economic structure. This permits the evaluation of the multiple paths in the project network. Also, the limiting values of the PNET transitional correlation (0,1) permits the estimation of bounds on all of the derived variables. Project cost and revenue are evaluated in terms of current, total and discounted dollars, thereby emphasizing the economic effects of time, inflation and interest on net present value and internal rate of return. The internal rate of return is evaluated from a variation of Hillier's method.
The analytical method is validated using Monte Carlo simulation. The validations
show that the analytical method is a comprehensive and extremely economical alternative to Monte Carlo simulation for economic risk quantification of large engineering
projects. In addition, they highlight the ability of the analytical method to go beyond the capabilities of simulation in the treatment of correlation, which are seen to be significant in the application problems. From these applications a technique to provide contingencies based on the probability of project success and to distribute the contingency to individual work packages is developed. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate
Identifer | oai:union.ndltd.org:UBC/oai:circle.library.ubc.ca:2429/30777 |
Date | January 1990 |
Creators | Ranasinghe, Kulatilaka Arthanayake Malik Kumar |
Publisher | University of British Columbia |
Source Sets | University of British Columbia |
Language | English |
Detected Language | English |
Type | Text, Thesis/Dissertation |
Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
Page generated in 0.0089 seconds