Comparison of the effects of prior information on the time of liner leak for selecting groundwater remediation design in a heterogeneous aquifer

Abstract

It is often necessary to use subjective information such that all effective information is used in decision making problems. The methods and key issues for the technical use of subjective information are analyzed. We present a technique for using subjective information about the time of liner leak for the planning of optimal groundwater remediation in an uncertain transmissivity field. The technique uses Monte Carlo simulation and Bayesian estimation theory. We also present applications of the technique to a fictitious problem. The results show that the prior distribution significantly affected optimal design planning. Therefore, the complete exclusion of subjective information may result in an unrealistic overestimate of uncertainty and hinder optimal planning.