Research on urban governance has for several years been dominated by such topics as democratization, public participation, and unequal power. A sub-domain in urban governance relates to the ‘digital age’ governance and urban data, and to managing a complex system of co-production and co-design with citizen participation as well as ensuring data literacy for efficient decision-making. Information technology offers tools to learn about preferences and usage, e.g., sensors, real-time monitoring, and spatial data analytics. Large datasets are often seen as assets for smart and efficient decision-making but rarely include or link to citizens’ attitudes, perceptions, and experiences. Finding patterns and trends of urban development becomes prime competencies for future urban decision-makers. This requires competencies of various kinds, including how to combine data sets and use state-of-the-art approaches to analyze the data.
We propose machine learning (ML) to advance and improve urban management, and foremost, urban decision-making. ML is a method to train algorithms, to interpret patterns in data and predict outcomes based on statistical analysis. The research project aims to (a) increase knowledge of how to use ML for efficient urban governance and decision-making, (b) identify mechanisms of urban change and development, and (c) encourage environmentally friendly choices among citizens. We thus aim to study how ML can be used in urban management and development, with the prime target of sharing information with stakeholders and increasing their competence and knowledge base on the topic of ML in urban governance. The following research questions guide the project:
1. Which data is relevant for effective decision-making in urban management?
2. Which data is suitable for extracting information for urban decision-making?
3. How can value models drive decision-making in urban management?
4. Which practices and competencies are required to use ML in urban management?

