Premium
Use of decision trees to value investigation strategies for soil pollution problems
Author(s) -
Okx J. P.,
Stein A.
Publication year - 2000
Publication title -
environmetrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.68
H-Index - 58
eISSN - 1099-095X
pISSN - 1180-4009
DOI - 10.1002/(sici)1099-095x(200005/06)11:3<315::aid-env411>3.0.co;2-0
Subject(s) - sampling (signal processing) , environmental remediation , sample (material) , probabilistic logic , decision tree , computer science , pollution , environmental science , risk analysis (engineering) , operations research , contamination , business , mathematics , data mining , ecology , chemistry , filter (signal processing) , chromatography , artificial intelligence , computer vision , biology
Remediation of a contaminated site usually requires costly actions, and several clean‐up and sampling strategies may have to be compared by those involved in the decision‐making process. In this paper several common environmental pollution problems have been addressed by using probabilistic decision trees. Decisions in this paper are on how to sample to detect a hot spot at different consumer's risks, whether to take additional samples to obtain a sufficiently precise estimate of an environmental contaminant, how to value environmental sampling strategies and how to choose between different test procedures prior to the remediation of soils. Decision trees combine costs with possible actions and chance events. For the studies analysed, they proved to be of use in making a well‐supported decision. In the case studies, we found that the value of surveys depends not only on the costs of the survey itself, but equally on the ratio of expected failure or success and the related costs. Thus, minimising costs on surveys and tests can lead to considerable losses in terms of effectiveness and efficiency. Copyright © 2000 John Wiley & Sons, Ltd.