Page 282 - Sustainable On-Site CHP Systems Design, Construction, and Operations

Managing Risks during CHP Plant Construction 255

sampling, but Latin Hypercube sampling is an alternative approach, where the
probability distribution is divided into intervals of equal probability. The approach
does provide increased accuracy at the expense of more computer time and memory
requirements and is beyond the scope of this chapter.
Finally, one should be prepared to interpret the results carefully. Remember to look
for any interdependence between the elemental categories. Contracting experience and
intuition are required when employing Monte Carlo simulations. The strength of
correlation between two variables will show the interdependence. Examine the shape
of the resultant distribution and the cumulative frequency diagram. The cumulative
frequency distribution allows one to examine the probability of obtaining a unit cost
below a chosen unit cost. Basically, the distribution allows one to consider the chance of
the most likely cost being achieved. Consider the statistics produced from the data, and,
when possible, take the time to test the sensitivity of the data obtained and assumed by
performing a sensitivity analysis on the key elements also used in the analysis.

References

Bowker, G. L., 1972, Engineering Statistics, 2d ed., Prentice-Hall, Inc., Englewood
Cliffs, NJ.
Flanagan, R. and Norman, G., 2003, Risk Management and Construction, Blackwell
Publishing, Oxford, UK.
Langley, R., 1971, Practical Statistics Simply Explained, Dover Publications, Inc.,
New York, NY.
Larson, R. and Farber B., 2006, Elementary Statistics, 3d ed., Pearson Prentice Hall, Upper
Saddle River, NJ.
Lifson, M., 1972, Decision and Risk Analysis for Practicing Engineers, Cahners Books,
Boston, MA.