American Society of Civil Engineers

Using Neural Network to Investigate the Environmental Impact of an Abandoned Landfill

by S. A. Mryyan, (Adjutant General’s Department, Environmental Compliance Manager, 131 SW 27th St., Topeka, KS 66611, E-mail: and Y. M. Najjar, (Professor, Kansas State University, 2118 Fiedler Hall, Civil Engineering Department, Manhattan, KS 66506 E-mail:
Section: Environmental Processes IV, pp. 1-10, (doi:

     Access full text
     Purchase Subscription
     Permissions for Reuse  

Document type: Conference Proceeding Paper
Part of: World Environmental and Water Resource Congress 2006: Examining the Confluence of Environmental and Water Concerns
Abstract: Advances in technology are allowing scientists and engineers to perform their jobs in ways that are both more time-efficient and cost-effective. With the number of abandoned landfills increasing every year, it is essential that the cost of investigation and remediation of such sites remain as low as possible. To achieve this objective, many site investigator teams are choosing to employ the powerful function approximation feature of artificial neural networks (ANN). Once the desired ANN profiling system is trained on existing data, the system can be used to produce efficient predictions for the unexplored regions. Similar ANN profiling systems were utilized herein in the investigation of an abandoned landfill site in Kansas City, KS. The developed ANN systems were trained on existing data and then were used to predict the amounts and distribution of arsenic and lead contaminates within the landfill area. The site investigating team was able to capitalize on the information generated via the developed ANN systems in order to determine the strategic locations for further testing. Therefore, less money was spent since no further testing was needed in the relatively uncontaminated regions.

ASCE Subject Headings:
Neural networks
Environmental issues