From Microsimulation to a Digital Twin of Society: Methodological and Data Foundations of project InnoTwin

Digital twins are increasingly regarded as a key technology for analysing complex systems. While the concept is well established in engineering and industry, its transfer to societal systems remains methodologically underdeveloped. This presentation discusses, using our BMFTR-founded project InnoTwin (www.innotwin.de) as a case study, what it scientifically means to construct a digital twin of society, and how this approach differs from classical microsimulation. InnoTwin is based on an agent-based model that explicitly represents individual life courses, behavioural responses, and social interactions, thereby moving beyond the analysis of average effects.

A defining feature of a digital twin is the continuous feedback loop with the real world. New empirical data are regularly used to update and recalibrate the model, while simulated policy scenarios generate testable expectations that can be compared to observed outcomes. Deviations between simulation and reality are used to iteratively refine behavioural rules and parameters. In this way, the model becomes an adaptive, data-driven representation that evolves in parallel with the society it describes.

Another focus of the contribution is on data-related challenges. Classical surveys and commuting studies are subject to systematic sampling biases that distort or underrepresent specific population groups. We show how synthetic populations can be used to correct such distortions and to generate coherent, robust microdata. On this basis, we argue that for certain research questions synthetic samples may yield more reliable inference than direct subsamples, without questioning the indispensable role of empirical surveys as training and calibration data for simulation models.

Using current applications on childcare expansion, labour markets, pensions, and long-term care, the presentation illustrates how a digital twin can serve as an experimental environment for analysing long-term policy effects. The contribution thus provides a methodological clarification of the distinction between microsimulation and a digital twin of society, and advances the role of synthetic data in evidence-based policy analysis.