It is often said in the field of complex systems that such systems achieve self-organisation through simple order-generating rules. Under what conditions is such self-organisation achieved?
I do not think that a complex system can attain self-organization without an external, probably continuous, help because there is a natural tendency in this universe for an increase in entropy (or more & more disorder). It is well-known that destruction is easier than construction because destruction causes more entropy "in full agreement with the laws of thermodynamics which have many dimensions besides chemistry & physics". In my opinion, self-organization occurs under one condition: an outside help for the object targeted for organization.
This is a good question. In addition to what @Kamal and @Nizar have incisively observed, there are a few more things to consider.
An overview of different types self-organizing systems is given in
J.A.C. Gomez, Self-organization in biology: From quasispecies to ecosystems, Ph.D. thesis, Unversidad Carlos III de Madrid, 2010:
http://matematicas.uc3m.es/jcapitan/tesisjoseacapitan.pdf
See Section 1.3 (Self-organization, universality and scaling), starting on page 6. The challenge is for any entity to survive (and flourish) when it finds itself in situations in between order and disorder. A very detailed introduction to self-organization in biological systems is given in Section 1.5, starting on page 8.
A very detailed introduction to conditions (and rules) required for self-organizating is given in
C. Gershenson, Design and control of self-organizing systems, Ph.D. thesis, Vrije Universiteit Brussel, 2007:
http://cogprints.org/5442/1/thesis.pdf
For example, in typical swarming rules, agents start to move with varying speeds towards the centre of a swarm, while mutually adjusting their velocities to avoid collisions (see Section 3.2, starting on page 24, especially page 27). This view of swarming behaviour carries over to artificial self-organizing systems such as a swarm of robots (see Section 3.5.1, starting on page 33) and self-organizing traffic lights (see Section 5.1, starting on page 62).
Self-organization canon and should not be taken as a general token. Several efforts have been made to explain and understand self-organization. Thus, f.i., it appears that boolean networks can be safely identifies as genesis of self-organization. Translated into biological terms, this refers to self-catalytic cycles, which are essentially open and increasing.
Furthermore, self-organization has been said to be the outcome of scale-free networks, This explanation focuses on a different scale, but has also proved solid and robust.
Get inspired : "Self-Organization in biological systems", http://press.princeton.edu/titles/7104.html
Self-organization can not work unless there is a outside interference to manage it.
Self-organization is the spontaneous often seemingly purposeful formation of spatial, temporal, spatiotemporal structures or functions in systems composed of few or many components. In physics, chemistry and biology self-organization occurs in open systems driven away from thermal equilibrium. The process of self-organization can be found in many other fields also, such as economy, sociology, medicine, technology.
http://www.scholarpedia.org/article/Self-organization
I would say quite the contrary. Self-organization is one of those marvelous concepts and approaches in complex science that steps far aside from causality. There is no need of an external cause for a self-organized system to emerge. This is what a boolean network entails.
Please allow me to put it straightforward in the jargon of complexity: a complex system appears with a minimum complexity. This means, a complex system is not the result of the aggregation of elements. Quite on the contrary, it is the outcome of a synthesis.
systems become self organizing if the internal entropy is decreasing, at least for physical systems. But they are not so complex
Very nice, indeed dear Guido. Even conspicuous physicists working on complex systems have already acknowledged that the physical systems are not as complex as the biological or the human systems. As we know, in fact, pearhead research in phsyics or math, f.i. has turned the focus on living systems in general. A most fascinating shift!
It is a process by which systems of many components tend to reach a particular state, or a small volume of their state space, with no external interference.
Dear Kamal,
What do you understand on the adjective standardised?
What do other participants mean on standardised?
It is possible that the meaning of this expression has got a variance of meanings.
Dear Mohamed,
I think the self-organisation must have various and hierarchical degrees which may highly influence the necessary conditions.
Dear Mohamed,
Does your question concern on natural or artificial systems?
Dear Andras,
Yes, my friend. I agree with you that the self-organisation must have various and hierarchical degrees which may highly influence the necessary conditions.
I agree with Dr Nizar that there is a natural tendency in this universe for an increase in entropy (or more & more disorder). This is also true for any isolated systems, but not always true for open systems.
Dear András, voilà a fine-tuned question. Honestly speaking, in the context of self-organizzation the language about "standards" is not possible at all. A number if studies have sufficiently show that self-organization is a structure or dynamics that occurs in far-from equilibrium conditions. Hence, no standards, no matter what. A further consequence is that it makes no sense to speak of inputs and outputs - not even in terms of asymmetry between them - when dealing with self-organized systems. This is one of the marvels of complex science, in fact.
Dear Mohamed,
In human complex systems I like the views of Csikszentmihalyi and Sawyer who identified conditions for autotelic experience (or flow) in indivuals and in groups. I have attached a case study of a group of 200 people that went through a complex self-organizing experience and then reflected on the conditions that made it happen. I juxtaposed them with Sawyers and Csikszentmihalyi's conditions so this short piece contains the references to their work as well.
Best, Sergej
Article Mise en Place 'in groove'
Lots of good answers have already given by my predecessors! Dear @Mohamed, I am attaching some more resources which are valuable for this thread! "Self-organization is another of the foundational concepts within complex systems that is proving particularly relevant to the world of the 21st century as we see collaborative self-organizing groups such as Wikipedia and the Linux foundation emerge. But self-organization is more than just a social phenomenon..." THE SCIENCE OF SELF ORGANIZATION AND ADAPTIVITY by Francis Heylighen - article is attached!
https://www.youtube.com/watch?v=BTR17I_Eb_o
http://www.elearning.fotonlabs.com/lessons/self-organization/
http://pespmc1.vub.ac.be/papers/EOLSS-Self-Organiz.pdf
There are lots of models with this behavior. I myself am working over consensus algorithms to reach agreements in agent.based systems based on the ideas of
http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4118472
http://personal.stevens.edu/~yguo1/EE631/Murrau04.pdf
This is not a new idea: Alan Turing worked on it about pattern formation and chemical reactions in 1952
http://www.dna.caltech.edu/courses/cs191/paperscs191/turing.pdf
@ Dr Kamal,
if the input and output of the system are standardised, do we still have a complex system? doesn't the system become complicated rather than complex?
@Mohamed sure! the complexity does not depend on how simple the input/output is. Indeed, very complex systems are modeled with simple equations/rules. Usually the complexity is related with the number of participants, how are they connected or chaotic evolution, among others.
Complex systems are MIMO systems, thus Multi Input Multi Output Systems, so their equations are not so simple!
In particular for human, project-oriented efforts
http://rstb.royalsocietypublishing.org/content/364/1518/781
I still like to focus on ants when I contemplate such matters.
The self-organizing exploratory pattern of the argentine ant
JL Deneubourg, S Aron, S Goss, JM Pasteels - Journal of insect behavior, 1990 - Springer
... the term self-organization, following an identical use in other natural systems (Nicolis and
Prigogine, 1977; Schieve ... modifies the following ant's probability of choosing left or right, a
positive-feedback system that, after ... The Self-Organizing Exploratory Pattern of the Argentine Ant ...
Cited by 796 Related articles All 17 versions Cite Save
[HTML] from princeton.edu
[BOOK] Self-organization in biological systems
S Camazine - 2003 - books.google.com
... In self-organizing systems, pattern forma- tion occurs through interactions internal to the system ...
The literature on nonlinear systems often mentions self-organization, emer- gent properties, and ...
CHAPTER 1-9 Figure 1.2 Figure 1.2 Self-organized patterns in biological systems ...
Cited by 2757 Related articles All 13 versions Cite Save
[PDF] from utexas.edu
Self-organization in social insects
E Bonabeau, G Theraulaz, JL Deneubourg… - Trends in Ecology & …, 1997 - Elsevier
... Task differentiation in Polisles wasp colonies: a model for self-organizing groups of robots. ... of
foraging activities in Pollutes dominulus Christ: a systemic approach to behavioural organization. ...
of worker foraging in paper wasps: A social feedback control system (Polistes fuscatus ...
Cited by 537 Related articles All 14 versions Cite Save
[PDF] from neu.edu
[PDF] Emergence of constraint in self-organizing systems
S Guerin, D Kunkle - Nonlinear Dynamics, Psychology, and Life …, 2004 - ccis.neu.edu
... Additionally, Francis Heylighen (2002) provides a good introduction to related concepts central
to self-organization. ... As systems self-organize, changes in Page 3. ... Thin agents are typically
associated with the modeling of complex adaptive systems and self-organizing systems. ...
Cited by 72 Related articles All 14 versions Cite Save More
[PDF] from persiangig.com
[BOOK] Ant Colony Optimization and Swarm Intelligence: 5th International Workshop, ANTS 2006, Brussels, Belgium, September 4-7, 2006, Proceedings
M Dorigo - 2006 - books.google.com
... we would like to thank all the people who helped in organizing ANTS 2006. ... Organization IX
Additional Referees Prasanna Balaprakash Leonora Bianchi Cecilia Di Chio Frederick ... 516
Muzaffer Kapanoglu, Fehime Utkan Self-organized and Social Models of Criminal Activity in ...
Cited by 5763 Related articles All 26 versions Cite Save
[PDF] from psu.edu
Entropy and self-organization in multi-agent systems
H Van Dyke Parunak, S Brueckner - Proceedings of the fifth international …, 2001 - dl.acm.org
... agents follow economic fields generated by myriad individual transactions, and self-organization
in the ... The movement of currency in such a system provides similar functions to those of ... of ordered
and disordered systems is ubiquitous in robust self-organizing systems, and that ...
Cited by 226 Related articles All 23 versions Cite Save
[PDF] from ulb.ac.be
Self-organized shortcuts in the Argentine ant
S Goss, S Aron, JL Deneubourg, JM Pasteels - Naturwissenschaften, 1989 - Springer
... Self-organized Shortcuts in the Argentine Ant ... ants comparing the differ- ent lengths of each branch,
but is in- stead a collective and self-organizing process, resulting ... also the principal factor behind
the organiza- tion of the collective exploration [5] and spatial organization [2, 13 ...
Cited by 819 Related articles All 11 versions Cite Save
[PDF] from researchgate.net
Self-organizing nest construction in ants: individual worker behaviour and the nest's dynamics
NR Franks, JL Deneubourg - Animal Behaviour, 1997 - Elsevier
... In self- organization the components interact locally with- out an overview of the pattern or the
process they are ... Franks & Deneubourg: Self-organizing nest construction 783 ... building There
is good evidence that Leptothorax workers use a template partly to organize their building ...
Cited by 134 Related articles All 11 versions Cite Save
[PDF] from arxiv.org
When can we call a system self-organizing?
C Gershenson, F Heylighen - Advances in Artificial Life, 2003 - Springer
... purpose of the system so that we can see the attractor as an “organized” state and to ... An ant colony
will not be self-organizing if we describe only the global behaviour of the ... We have to remember
that the description of self- organization is partially, but strongly, dependent on the ...
Cited by 164 Related articles All 19 versions Cite Save
[PDF] from ucla.edu
Self-organization in sensor networks
TC Collier, C Taylor - Journal of Parallel and Distributed Computing, 2004 - Elsevier
... Self-Organization in Biological SystemsPrinceton University Press, Princeton, NJ (2001). [7];
GD Caro, M Dorigo; ... Ant colony system: A cooperative learning approach to the traveling salesman
problem. IEEE Trans. Evol. ... Convergence properties of self-organizing maps. ...
Cited by 98 Related articles All 8 versions Cite Save
Most answers are about self-organization in biology. But some economists also describe those phenomena (although this is not a mainstream).
I can mention Alan Kirman. I know that he studied self-organization of ants and fish market in Marseille, and here I just found his recent publication criticizing principles of classical economics: http://www.paecon.net/PAEReview/issue64/HelbingKirman64.pdf
Brian Arthur is another economist interested in self-organization. See for example his book "Increasing returns and path dependence in the economy".
One thought or question I have currently is "To what degree do we need to talk about all key social concepts when we determine something like the model suggested above in "by Complex systems achieve self-organisation if the entrance (income) and exit (outcome) of the system are standardized"?
This means that Social Constructs making up these two (income and outcomes) will need to be observed and acknowledged, doesn't it?
Interesting question, it is my thesis research question actually. I focus my research on human systems, in which I found the following principles necessary so far to achieve self-organizing behavior;
1. Distribution of control/authority.
Distribution of authority makes the system highly complex, and, therefore, self-organizing, because the freedom of each agent is increased, which creates a chaotic context to work with. The chaos and tensions between all agents because of the distributed control in an organisation creates the fuel for novelty. New pathways are continuously explored and tested, first using mental models, then testing it out in the real to see which path to take is the best 'fit' based on external tensions the system has with other systems.The chaos caused by the competition and cooperation of the agents in the organization is, therefore, necessary to put the system far from equilibrium, on the edge of chaos, so that the whole organization can continuously evolve according to its purpose. Which brings me to the following principle;
2. Evolutionary purpose/goal.
All agents in the organization are constantly (re)defining the purpose of the organization based upon the information they have of their surroundings. The existence of the organization is rooted in evolutionary principles, so the organization has to make itself relevant to other systems in their environment in order to survive and thrive. This purpose is formed bottom up by the agents, and creates a structure/a higher order within which the agents can execute their freedoms. This structure influences the behavior of the agents by top-down causality through positive and negative feedback loops. Decisions by individual agents that add to this purpose are enforced through positive feedback loops, deviant behavior that deflects this purpose are certainly allowed to a certain level in order to experiment, but probably corrected or cancelled because of the negative feedback loops if it turns out that the behavior is not contributing or damaging the purpose of the organization.
3. Qualities of the people/agents in the organization.
In order for self-organization to work, the people in the organization need to be willing to take responsibility, initiative and be willing to make mistakes. They have to differentiate themselves from others in their professions and expertise in order to create diversity in the system, plus they have to be able to communicate and exchange clearly with other agents so that there is a high level of connectivity between all agents. Basically said, there has to be diversity and connectivity between the agents to create a complex enough context out of which novelty can occur.
(4.) Self-organizing systems need constant energy to sustain its processes based on the principles of thermodynamics.
More obvious, following evolutionary principles, an organization of people are just like biological systems, semi-autonomous and requires external help/inputs to function. Organizations are co-dependent upon relevant systems in their environments. No organization is a one-man island on the longer term. So they require energy, matter or information (raw materials of a low level of entropy) to put products and services back into their environment to create a level of reciprocity. Waste is dissipated into 'buffer zones' into neighboring systems with a higher level of entropy. This dissipative structure of self-organizing systems is key in creating efficiency and prolonging the system from maximum entropy, until eventually, it ceases to exist, like pretty much everything.
This in short is what I can conclude,
Happy to hear thoughts about it.
Kees
Dear @Mohamed, check the following presentation. It is valuable resource.
Complex Adaptive Systems: Emergence and Self-Organization
https://www3.nd.edu/~gmadey/Activities/CAS-Briefing.pdf