Welcome to the site of

Manfred Füllsack Univ. Prof. Dr., Systems scientist at SIS-logo

Fuelli Mail to: manfred.fuellsack (at) uni-graz.at
Merangasse 18/I A-8010 Graz, Austria
T +43-316 380-3235
F +43-316 380-9585

Complex Systems, Networks, Games, Systems modeling, Social economic and ecological aspects of work



Georg Jäger Ass.Prof. Dr.rer.nat. BSc MSc



Computational Systems Science
Systems modeling
Simulation of complex systems

Homepage: http://jaeger-ge.org

Marie Kapeller, Dr.rer.nat. BSc MSc


marie.kapeller (at) uni-graz.at

Cooperation research
Systems Modeling

Annelies De Meulenaere, BSc MSc


annelies.de-meulenaere (at) uni-graz.at

Supply chain optimisation
Computational Systems Science

Simon Plakolb, BSc MSc


simon.plakolb (at) uni-graz.at

Traffic simulations
Computational Systems Science
Systems modeling
Functional programming

Raven Adam, BSc MSc


raven.adam (at) uni-graz.at

Machine learning
Text mining
Climate physics

Daniel Reisinger, BSc MSc


daniel.reisinger (at) uni-graz.at

Agent based modeling
Behavior research
Systems modeling

Laura Zilian, BSc MSc


laura.zilian (at) uni-graz.at

Labor market modeling

Doris Prach, BSc


doris.prach (at) uni-graz.at


A system is "a concept that hardly ever aims at less than the whole. At times though deployed just to delimit objects for investigation, the term per definition obliges to consider interrelations, and interrelations tend to reach beyond any limits. They are never purely internal. The concept hence pushes its own boundaries. It forces to consider a system’s context and thereby calls to mind that any analytical containment called system will always be a system in systems. No final delimitation will stand. So (...), if you decide to go for systems you have to go all the way.

Analysis however, would be lost in a limitless space. Where should scientific observation begin and where should it end? Hence, as other scientific objects, systems too must be – at least temporarily – limited in order to function as analytical means. But if they are delimited anyway, you may object, what then, apart from their name, distinguishes systems from conventional objects as they are delimited in sciences all along?

The difference that makes a difference in this case (...) is the digital computer. Systems sciences deploy digital machines in their own right as analytical tools, which, as we all know, allow for an unprecedented synopsis of interrelations even though they are limited in the sense of having to start at an externally predefined level of order (...). In other words, computers enable investigation into a complexity of relations and interactions that was not conceivable before the digital age, and they do so in spite of their technical limitations. They allow “crawling the micro-causal web”, as Mark Bedau (1997) called it, a web of, albeit incomplete, but still illustrative interactions of lower level components that by themselves do not show the properties of the higher level phenomena to which their aggregations emerge.

With this, the computer allows for the verification of insights, of which previously the possibility could be suspected at best, but could never be proofed. The computational power of contemporary digital machines indeed provides ground for “a new kind of science” (Wolfram 2002). And in this kind of science (...) it does not matter that the machine so far is not (yet) able to simulate emergences that span much more than a couple levels of order (...). Like in all of sciences, it is the extrapolation that matters. It is the insight (never the proof) that a local interaction of autonomous agents with noisy and incomplete information can cause higher order phenomena like solving complex tasks that require the integration, computation and exchange of information amongst an entire population of agents. It is the model that in its abstractness and incompleteness suffices to illustrate how highly complex phenomena can be caused by simple mechanisms that by themselves would not be suspected of having far-reaching effects. Or in still other words, it is the maybe abstract, incomplete and noisy model that in interaction with other such models might give rise to insights that none of them could have triggered on their own."

Excerpt from: Füllsack 2013.

Bedau M.A. (1997) Weak emergence. Philosophical Perspectives 11: 375–399.

Wolfram S. (2002) A New Kind of Science. Wolfram-Science.

Füllsack M. (2013) Systems Sciences and the Limitations of Computer Models of constructivist Processes. Constructivist Foundation 9(1): 33.

Master theses topics (if interested please apply):
Digitally assisted self-organization in labor relations
Commons, cooperation and networks
Assessing productivity in circular economics
Early Warning Signals
Data-based ABM
Data mining and sustainability

A (partially) interactive introduction to Systems Sciences

Information on topics covered in 2019/20-lectures

Data in Systems Sciences - Script

USW Computational Basics - IPython-Notebook als Skriptum für die VO 'USW Computational Basics' (in German)

Einführung in Python - PDF-Skriptum für die VO 'USW Computational Basics' (in German)

  • Hosted by University of Graz