How many refugees per municipality? Project from Lower Saxony aims to help

The distribution of refugees across German municipalities is often not very targeted, say researchers at the University of Hildesheim. Their approach will improve this.

Lower Saxony – How can the distribution of refugees to municipalities in Germany be made more efficient and fairer? Researchers at the University of Hildesheim in Lower Saxony have been investigating this question. To do this, they have developed software that uses an algorithm to query the needs and offers on both sides: those of those seeking protection and those of the municipalities.

They called the prototype of this algorithm “Match’In”. It is intended to ensure that the opportunities for integration are better exploited and that this is therefore a benefit for the refugees themselves and for the local society. But how does it work exactly?

Distribution of refugees according to need – algorithm should help

The algorithm is used to systematically record and compare the individual requirements of those seeking protection and the structures in the municipalities. According to the current procedure, the decision as to which municipality people seeking protection are assigned to as part of their asylum procedure is not a self-determined one for either the people concerned or the receiving municipalities – although it is extremely far-reaching, since people usually stay in this place for a longer period of time after often long escape histories and stays in reception facilities.

However, whether this place is also suitable for them has not yet been sufficiently considered, according to a statement from the University of Hildesheim, which carried out the project in cooperation with the FAU Erlangen-Nuremberg.

Project from Lower Saxony helps promote the integration of asylum seekers

The distribution regulations made at the state level are primarily based on admission quotas, not on individual needs or the resources of the municipalities. “This means that people are assigned to municipalities where they cannot find suitable offers or do not have the opportunity to contribute their own skills,” explains project manager Danielle Kasparick.

This often leads not only to dissatisfaction on both sides, but also to a move to another location as soon as this is legally possible. This means that important initial integration successes are lost, which also entails costs for the municipalities. “We want to counteract these avoidable ‘mismatches’ with Match’In,” says the project manager. The aim of the project was therefore to develop an algorithmic process that is scientifically sound and at the same time suitable for practical use. To this end, relevant matching criteria were developed, systematically collected and included in the distribution.

Software determines profiles of refugees and compares them with the offers of the municipalities

This innovative approach resulted in a tailor-made software tool that municipalities can use to create a detailed profile. In the initial reception centers of the federal states, the needs and skills of those seeking protection are first determined, as are the refugees’ professional qualifications and health needs.

The specially developed “Match’In algorithm” compares this information with the offers of the municipalities, for example open jobs or care options, and generates distribution suggestions based on this. These suggestions offer the employees of the allocation offices valuable insights into how well the various municipalities match the individual profiles of those seeking protection.

Although the final decision is facilitated by technical support, it remains in the hands of the employees who are best able to assess the situation on site. This ensures that the integration is not only efficient, but also humane and needs-based.

“Procedure can be successfully applied” – project from Lower Saxony is still ongoing

The algorithm prototype will be in use in the participating municipalities until autumn 2024. The scientific project team spent three years working on this task and also monitored and evaluated the practical implementation. “At the end of the project period, it is clear that the matching process can be successfully applied in the different federal states,” concludes Hannes Schammann from the University of Hildesheim.

But the end of the planned project period does not represent the completion of the project, explains Petra Bendel, project manager at FAU Erlangen-Nuremberg. Rather, they want to examine how the distribution process can be further developed in a meaningful way in regular operation. “Match’In” offers an important impetus for this.