Skip to main content

In recent years, decision analysis has become an important technique in many disciplines. It provides a methodology for rational decision making allowing for uncertainties in the outcome of several possible actions to be undertaken.

TitleOptimal decisions that reduce effects of sewer systems on receiving waters : Bayesian decision analysis as a tool for defining monitoring needs
Publication TypeMiscellaneous
Year of Publication2001
AuthorsKorving, H, Clemens, F
Pagination8 p. : 8 fig.
Date Published2001-10-01
PublisherDelft University of Technology, Department of Environmental Engineering
Place PublishedDelft, The Netherlands
Keywordsbayesian networks, monitoring, sdisan, sdiurb, sewers, uebw, urban areas
Abstract

In recent years, decision analysis has become an important technique in many disciplines. It provides a methodology for rational decision making allowing for uncertainties in the outcome of several possible actions to be undertaken. An example in urban drainage is the situation in which an engineer has to decide upon a major reconstruction of a system in order to prevent pollution of receiving waters due to CSOs. This paper describes the possibilities of Bayesian decision making in urban drainage. In particular, the utility of monitoring prior to deciding on the reconstruction of a sewer system to reduce CSO emissions is studied. Our concern is with deciding whether a price should be paid for new information and which source of information is the best choice given the expected uncertainties in
the outcome. The influence of specific uncertainties (sewer system data and model parameters) on the probability of CSO volumes is shown to be significant. Using the Bayes' rule, to combine prior impressions with new observations, reduces the risks linked with the planning of sewer system reconstructions. (Authors' abstract)

Notes8 ref.

Disclaimer

The copyright of the documents on this site remains with the original publishers. The documents may therefore not be redistributed commercially without the permission of the original publishers.

Back to
the top