Hypernetworks in the Science of Complex Systems Part I. 1 st PhD School on Mathematical Modelling of Complex Systems July 2011, Patras, Greece

Transcription

1 Hypernetworks in the Science of Complex Systems Part I

2 Hypernetworks in the Science of Complex Systems I Complex Social Systems science necessarily involves policy

3 Hypernetworks in the Science of Complex Systems I Complex Social Systems science necessarily involves policy policy is designing the future and managing the present

4 Hypernetworks in the Science of Complex Systems I Complex Social Systems science necessarily involves policy policy is designing the future and managing the present involves language, logic and mathematics

5 Hypernetworks in the Science of Complex Systems I Complex Social Systems science necessarily involves policy policy is designing the future and managing the present involves language, logic and mathematics Policy Science Design & management

6 Hypernetworks in the Science of Complex Systems I Complex Social Systems science necessarily involves policy policy is designing the future and managing the present involves language, logic and mathematics Policy Science Design & management II Hypernetworks generalise networks from 2 to many dimensions can model multilevel system dynamics support design and management of complex systems

7 We live in a complex & dangerous world

8 We live in a complex & dangerous world run by powerful people?

9 We live in a complex & dangerous world addressing complex problems

10 We live in a complex & dangerous world claiming successes

11 We live in a complex & dangerous world but with many failures

12 We live in a complex & dangerous world but with many failures

13 We live in a complex & dangerous world but with many failures We need new understanding of the social and technical world We need new understanding of how it works and how it can be managed We need a science of complex systems We need many thousands of people trained in this science

14 We live in a complex & dangerous world but with many failures We need new understanding of the social and technical world We need new understanding of how it works and how it can be managed We need a science of complex systems We need many thousands of people trained in this science The world can be a better place

15 Emerging Perspectives on Complex Systems Domains physics chemistry biology psychology economics robotics computing Deep special domain knowledge requires specialists

16 Emerging Perspectives on Complex Systems Domains physics chemistry biology psychology economics robotics computing Methods Mathematics Statistics Stat Physics Computing Data handling Deep special domain knowledge requires specialists Complex systems science cuts across the domains

17 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves, e.g. - biological evolution - political revolution - manufacturing done in China - changes in women s fashions - scientific paradigm shifts - economic meltdown often adapting to unpredictable changes in their environments

18 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves System Environment often adapting to unpredictable changes in their environments

19 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves System Environment often adapting to unpredictable changes in their environments

20 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves System Environment often adapting to unpredictable changes in their environments

21 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves System Environment often adapting to unpredictable changes in their environments

22 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves System Environment often adapting to unpredictable changes in their environments

23 Emerging Perspectives on Complex Systems Complex systems are able to completely reconfigure themselves co-evolution with their environment System Environment often adapting to unpredictable changes in their environments

24 Emerging Perspectives on Complex Systems Characteristics of complexity Adapt to changes in their environment Sensitivity to initial conditions (almost all human systems!) Systems of systems of systems many heterogeneous parts ill-defined subsystem boundaries macro emerges from micro interactions of parts bottom-up and top-down dynamics (not understood) Etc.

25 Emerging Perspectives on Complex Systems Characteristics of complexity Adapt to changes in their environment Sensitivity to initial conditions (almost all human systems!) Systems of systems of systems Etc. many heterogeneous parts ill-defined subsystem boundaries 10 challenge How many objects are in this room? macro emerges from micro interactions of parts bottom-up and top-down dynamics (not understood)

26 Emerging Perspectives on Complex Systems Embracing complexity in design System involving human beings are artificial systems Systems as they ought to be. The are designed drugs, cities, waterways, buildings, power stations

27 Complexity and Design? Designed systems can be complex drugs, cities, armies, buildings, power stations

28 Complexity and Design? Designed systems can be complex drugs, cities, armies, buildings, power stations Design processes can be complex chip fabrication, factories, supply chains,

29 Complexity and Design? Designed systems can be complex drugs, cities, waterways, buildings, power stations Design processes can be complex chip fabrication, factories, supply chains, Design environments can be complex regulation, fashion, bank rate, exchange rates,

30 Complexity and Design? Designed systems can be complex drugs, cities, waterways, buildings, power stations Design processes can be complex chip fabrication, factories, supply chains, Design environments can be complex regulation, fashion, bank rate, exchange rates, Design is a complex cognitive (social) process creating, learning, sketching, communicating,

31 Design establish requirements generate evaluate

32 Design change requirements establish requirements generate evaluate

33 Design change requirements establish requirements generate evaluate (generate evaluate) (change requirements) double cycle implies co-evolution!

34 Design The system hypothetical subsystems words, sketches, drawings, numbers equations assemble real stuff emergent properties stuff Design is a way of evolving models of systems

35 Design The system fully instantiated hypothetical subsystems words, sketches, drawings, numbers equations assemble real stuff emergent properties stuff blueprint Design is a way of evolving models of systems

36 Policy?

37 Policy? Most policy is an experiment

38 Policy? Most policy is an experiment but not scientifically instrumented Scientists cannot do these experiments they don t have the mandate the don t have the money Policy is designing the future

39 Policy? Most policy is an experiment but not scientifically instrumented Exception Helbing s work shows how good science can be applied to social systems

40 Policy as designing the future What sort of future is required? Sketch out possible futures Evaluate possible the futures Decide on a target future implement! Design & Policy involve prediction

41 Policy as designing the future System at t System at t+1 What is a prediction? How an it be tested?

42 Policy as designing the future What is a prediction?

43 Policy as designing the future What is a prediction?

44 Policy as designing the future What is a prediction? Will the designed system work?

45 Policy as designing the future Complex systems are investigated by simulation e.g. Agent-based modelling Prediction horizon Possible futures

46 Policy as designing the future Complex systems are investigated by simulation e.g. Agent-based modelling Prediction horizon Possible futures

47 Policy as designing the future is entangled with complexity science and design Policy Science Design & management

48 Policy as designing the future entangled with science and design scientific propositions policy propositions

49 Policy as designing the future entangled with science and design scientific propositions logical argument, data policy propositions rhetoric anything goes to win

50 Policy as designing the future entangled with science and design science-based policy problem which version of True? scientific propositions logical argument, data policy propositions rhetoric anything goes to win

51 Policy as designing the future entangled with science and design science-based policy problem which version of True? simulation? Conviction? scientific propositions logical argument, data policy propositions rhetoric anything goes to win

52 Policy & Design must play major roles in the science of complex systems Policy Science Design There is a power struggle for the Truth We need to understand better how models fit into the logic of policy We must meta-model ourselves inside the system

53 Experiments 1 How many objects in the room? 2 Fill in the questionnaires 3 Students suggest new items for the questionnaires 4 Pass back the questionnaires ladies forward or to the left gentlemen backward or to the right