This research investigates how a Spatial Decision Support System (SDSS) that is dedicated to communicate flood risk information can be analytically and communicatively supportive for laymen in flood prone areas that have varying preferences for communication methods. To divide residents into groups, the cultural theory of risk is used, by which residents’ perspectives can be classified into fatalist, hierarchist, individualist and egalitarian. Floodlabel.net, a prototype SDSS that aims to inform residents about their personal flood risk, is used as a case. This research concluded that improvements for floodlabel.net regarding the analytical and communicative support could be beneficial for bringing residents in general and residents from a specific group of cultural theory to action. Yet, a platform such as floodlabel.net should always be assisted by other communication methods for an optimal flood risk communication.
The broken window theory is one of the most influential, well-documented, and controversial perspectives in criminology. This study aimed to address a lack of a large-scale spatial data analysis. Police complaints and service requests were used to predict serious street crime through perceived disorder, to find the best trade-off between model performance and granularity for different spatio-temporal scales, and best performing machine learning model types. It provided a 3D Kriging random forest model for monthly tabular area crime numbers recording an R2 of 93,0% and MAPE accuracy of 81,0% for out-of-bag data. The best predictors were instances of social disorder.
The aim of my thesis was therefore to identify which spatial, infrastructural, consumer and sociocultural factors explain spatial variation in electricity, natural gas and water consumption by households within the municipality of Amsterdam. The results illustrate that twelve different factors underlie electricity, natural gas and water consumption. The factors building size, building type, income level and household size predominantly explain spatial variation in electricity consumption by households, whereas the factors building age, presence of district heating and income level are most important for explaining differences in natural gas consumption. Finally, the factors building type, household size and migration history were discovered to be most important for explaining spatial variation in household water consumption.
While crisis management and Spatial Data Infrastructure [SDI] assessment are both extensively studied, the link between the two fields is uncommon. This study combined literature of the two fields to clarify the role of SDIs within crisis management, and to set up a framework for SDI assessment. Conventional SDI assessment frameworks have been adjusted to the context of crisis management and the decision-making of the end-users, interviews have been conducted to see what needed to be finetuned. Crisis management is a complex and dynamic field which requires a different approach of SDI assessment compared to conventional SDI assessment.
Danny van Hienen
Web GIS assessments from the user perspective differ, because of the varying product characteristics, user types and user goals. Assessment developers have to invent the wheel repeatedly. A standardized method is lacking. This study designed and tested a step-by-step approach to give support at assessing web GIS from the user perspective. It includes support to define the research set up, define the questionnaire content, create an online questionnaire, and generate and interpret the results. The method is ready for application on a much wider scale and developing to a standardized guidance for anyone who wants to assess a web GIS.
In the last couple years there has been a large increase in number of malaria cases in Rwanda despite several investments. Gaining more knowledge about how the environment can influence the spatial and temporal distribution of mosquitoes can help to better understand malaria transmission patterns. This study, explored the spatial and temporal distribution of mosquitoes in the Ruhuha sector, Rwanda. A citizen science approach was used in combination with GIS to contribute to future researches in low-resource settings. Results showed that in the south there are more mosquitoes, which could be related to local environmental variables.
The aim of this research is to accurately classify urban land cover of the Netherlands. This is done by developing a novel geo-computational workflow that is able to automatically classify urban land cover on a large scale using high-resolution street view imagery and deep learning.
The performance of the trained models was measured in terms of accuracy, recall, precision and F1-score. The application of the workflow resulted in an overall accuracy of 52%. The performances achieved by the developed geo-computational workflow appeared to be not sufficient enough for urban land cover classification in a real-world application. In conclusion, future research should focus on improving the model that is developed in this study by increasing the number of training images for all used distances.
Electric Vehicles (EVs) are an opportunity for governments to reduce the greenhouse gas emissions and to improve air quality. There is an increasing need to understand the processes behind EV development and use. For this research, a behavioural model is developed to explore the behaviour of electric motorists in relation to charging point placement. The resulting simulation model includes behavioural concepts found in literature. Model verification, validation and scenario testing showed that the model works as designed and it has clear potential for testing real-life scenarios and using it for policy making. When more extensive datasets are retrieved for validation and calibration, the quality of the model will increase.