An Integrated MATLAB-ArcGIS Toolbox for Landfill Suitability Mapping Using Neural Networks

The Neural Network Toolbox for Landfill Suitability Mapping presents a novel approach to enhance solid waste management by automating landfill site selection using a combination of Python-ArcGIS and MATLAB environments. This study addresses the critical need for efficient landfill suitability mapping, which is essential for minimizing environmental risks and optimizing land use. We developed an automated toolbox that integrates the capabilities of MATLAB for neural network modeling with Python scripting in ArcGIS, facilitating a user-friendly workflow for decision-makers. Our methodology was tested in the northern region of Peninsular Malaysia, employing 14 specific criteria to generate training and testing datasets. The results of neural network model achieving an accuracy of 82%. This toolbox streamlines the process by automating various stages, from data preparation to suitability mapping, reducing the likelihood of human errors and minimizing processing time. The flexibility, interoperability, and user-friendly interface of the toolbox make it accessible to users with varying levels of GIS expertise, ensuring its adaptability to different geographic regions and environmental conditions. This research contributes to more efficient, accurate, and sustainable landfill site selection, benefiting decision-makers and supporting broader environmental protection efforts.