A decision support system for wet soil cleaning.

Abstract

Newly developed software can predict the essential characteristics of the sand product resulting from the ex-situ washing of contaminated soil, dredging sludge or breaker sand. The system was designed for the Dutch situation and it works on the basis of readily available information. The primary output of the system is an assessment of the reusability of the sand product, which is obtained by comparing the prediction of the residual contamination and engineering quality with accepted levels and standards for building materials. The system also provides a ranking of potential applications for the sand with respect to treatment cost, the amount of recycled material or the environmental quality of the product. The software integrates soil cleaning expertise and unit process modelling in a way that allows varying levels of refinement of input data, ranging from a simple identification of the pollution source to a detailed particle size distribution of the contaminated soil. Tests on a database containing information on 117 processed batches of soil, dredging sludge and breaker sand show an 80% success rate in predicting cleanability, even though the input contamination levels were predominantly taken from (relatively inaccurate) in-situ data.