Order Amid Uncertainty (Available in Geoenvironmental special issue only)

by James D. Englehardt, P.E., Assoc. Prof. of Envir. Engrg.; Univ. of Miami, Coral Gables, FL,
Ted W. Simon, Toxicologist; U.S. Environmental Protection Agency, Atlanta, GA,

Serial Information: Civil Engineering—ASCE, 1999, Vol. 69, Issue 6, Pg. 8A-13A

Document Type: Feature article


Bayesian statistical methods allow engineers to deduce probabilities from incomplete information. If there is a theoretical or empirical reason why a particular probability distribution will apply to a given variable, engineers can use that (wider) distribution to predict an outcome. The method is particularly helpful to environmental engineers trying to determine the outcomes of various cleanup strategies when some facts are missing, and usually it leads to a conservative risk assessment. Bayesian models have already been used successfully for engineering and construction projects, most notably to evaluate proposed revisions to the South Florida building code that were proposed after Hurricane Andrew, and to create a model for predicting the oil spill consequences of proposed changes in the Gulf of Mexico oil-transportation network.

Subject Headings: Probability distribution | Probability | Hazardous materials spills | Construction engineering | Bayesian analysis | Uncertainty principles | Statistics

Services: Buy this book/Buy this article


Return to search