Data on the number of floors is required for a variety of applications, ranging from energy demand estimation to flood response plans. Despite this, open data on the number of floors is currently not available at a nationwide level in the Netherlands. This means that it must be inferred from other available data. Automatic methods usually involve dividing the estimated height of a building by an assumed storey height. In some cases, this simple approach limits the accuracy of the results. Therefore, the goal of this thesis is to develop an alternative method to automatically infer the number of floors.

Three different machine learning algorithms are tested and compared: Random Forest, Gradient Boosting and Support Vector Regression. These algorithms are trained using data on the number of floors obtained from four municipalities in the Netherlands. In addition, 25 features are derived from cadastral attributes, building geometry and neighbourhood census data. These features are tested in different combinations in order to determine whether a specific subset yielded better results. Furthermore, a comparison is made between features derived from 3D building models at different levels of detail.

The results show that building height, particularly 70th percentile height, is most related to the number of floors. Other 3D geometric features are also found to be quite closely related to the number of floors, specifically roof area and volume. However, a higher level of detail did not improve the results. Cadastral features are also found to be relevant; mainly net internal area and, to a lesser extent, construction year. Furthermore, models based on a combination of different features performed better than models based on single categories of features.

The best predictive model achieved an accuracy of 94.5% and a Mean Absolute Error (MAE) of 0.06 for buildings with 5 floors or less. This represented a substantial improvement on the results of the geometric approach, which had an accuracy of 69.9% and MAE of 0.31. However, above 5 floors, model performance was substantially lower. Machine learning provided only a slight improvement on the geometric approach for these buildings. In this case, the best model had an accuracy of 52.3% and MAE of 0.62, whereas the geometric approach was 47.5% accurate and had a MAE of 0.70. A comparison of the cumulative error distributions showed that the best model mainly improved the fraction of buildings that were predicted with an error of less than 1 floor. Overall, these results show that machine learning partially provided a better estimate of the number of floors than a purely geometric approach.

Open dumping, open burning and burying of municipal solid waste (MSW) can be the cause environmental and public health issues. These practices are more prevalent in developing countries such as Mexico,where proper waste management systems are not present. Considering the environmental and health issues, it is therefore important to minimise the number of open dumps in Mexico. The construction ofsanitary landfills is regarded as the best alternative to open dumping since it is the a cost-effective and environmentally friendly solution.

An important part of constructing sanitary landfills is the selection of potential locations for these wastefacilities where investment will be made to build them. In order to select these locations first the weakspots need to be located. Weak spots are areas that do not have enough (proper) waste managementservices. Since Mexico does not have a national solid waste information system, a method to locate theseweak spots needs to be developed. With the use of the weak spots a method can be developed to select the potential locations for sanitary landfills that also takes the social, economical and legal constraintsinto account. The following research question is formulated: What are the weak spots in the current waste infrastructure network in Mexico and, based on this, where should strategic investment be madeto improve waste disposal? By answering this question, information will be provided on the issues withthe management of waste in Mexico with a focus on the areas of the weak spots and the locations where investment can be made to develop new sanitary landfills.

To detect the weak spots, a set of factors of different scenarios were developed, scored, overlaid, and visualised in maps. Regions that have the lowest score were detected as weak spots. To select the potential locations for investment in new sanitary landfills a spatial decision support system (SDSS) was developed and implemented as a QGIS plugin. The weak spots that corresponded to urban areas were used for analysis in the SDSS. This is due to the fact that it is more economically beneficial to construct sanitary landfills in urban areas.

The weak spot analysis showed that the southern region of Mexico, especially the state of Oaxaca, hadthe highest deficiencies in waste infrastructure. With the output from the QGIS SDSS plugin we are able to determine potential areas for new sanitary landfills in an automated manner.

This research has resulted in the visualisation of the weak spots in the Mexican waste infrastructure and the selection of potential locations where investment can be made for the construction of new sanitary landfills. The approach for locating the weak spots of the waste infrastructure can be used to find the weak spots in other types of infrastructure on a state and country scale in Mexico. The QGIS SDSS plugin could also be used to locate sanitary landfills in Mexico that violate the standards and regulations. The approach used to develop methods to detect the weak spots in the waste infrastructure and select potential locations for investment into new sanitary landfills could be used as a model for other countries to develop their specific approaches.