This thesis presents the construction of the second population-scale social multi-layer network and the first on Swedish and longitudinal registry data. Our study focuses on constructing a detailed multi-layer network that captures various aspects of possible social interactions within Sweden. We analyse the network's structure and dynamics to investigate its quality in measuring segregation. Regarding segregation, our case study examines five Swedish cities,with a particular focus on two cities experiencing significant social dumping, as reported by local news. Our findings suggest that the multi-layer network offers valuable insights into segregation patterns, highlighting its utility for social science research and policy-making. However, it also showcases the challenges when dealing with large scale multi-layer networks and the lack of feasible libraries and segregation measures. Therefore, this work contributes to the growing body of knowledge on social networks and segregation, setting a precedent for future longitudinal social network studies.