Current thesis indicates the outcomes of developing a series of data-driven researches aggregating remote sensing satellite imagery, machine learning tools, and GIS network analysis. In the first part of this study, a GIS-Based method is developed that considers spatial, environmental, and economic constraints using study regions derived from NTL imagery for a 40-km buffer around Regina, Saskatchewan, Canada. Results showed that road network stands as the most decisive factor for detecting suitable sites for landfills followed by protected areas. LST and NDVI fluctuations in the vicinity of 8 Canadian engineered landfills were evaluated to identify biothermal zones and environmental footprints of landfills in the second part of the study. Decreasing LST trend is found for forest lands near the landfills in 5 out of 7 landfills. In addition, monitoring LST changes were appeared more appropriate for identifying bio-thermal zones in landfills. Third part of the study were dedicated to develop a RF algorithm as a machine learning tool to identify landfill emission hotspots. It is believed that the incorporation of RF algorithm with remote sensing imagery can increase the spatial and temporal resolution for monitoring purposes. Results showed that the predicted LST, as a proxy for methane hotspots, agrees well with LST estimated from the conventional approach using Landsat-8 imagery, with R> 0.96. NTL imagery, vectors files, and MCDM tools for identification of the potential IDS in vulnerability due to IDS. Highway length is found as the most decisive factor on IDS probability among all classes, with membership grades ranging from 0.99 to 0.55. Fifth part of the study were dedicated to develop an analytical framework to optimize population coverage by landfills using network analysis and satellite imagery. Results showed that landfill regionalization index, (an indicator for covered population centers), is generally increased in proposed method compared to the status quo using two different truck travel time. The separation distance between the generation and disposal sites and truck capacity appears not a decisive factor in the optimization process. Sixth part of the study is developed as a continuation of the previous part with different optimization objectives focusing on maximizing accessibility of landfills (ACC) and minimizing the number of landfills (LMI). The results suggest that the accessibility was able to be improved between 15.4 and 794.7%, and the optimization methods were always effective in providing higher accessibility to residents. Improvements in the LMI yielded less improvement, ranging from -13.9% to 422.2%. Furthermore, correlation analysis showed that road length is one of the most important optimization parameters, and an intensified road network is more efficient. This finding may impact Indigenous communities who suffer from a lack of road network. Keywords: Waste management, landfills, illegal disposal sites, remote sensing, satellite imagery, MCDM tools, GIS network analysis, solid waste collection optimization, Random Forest model, machine learning