Divisão de Observação da Terra e Geoinformática

Essa página está em modo somente de leitura. Você pode visualizar a fonte, mas não alterá-la. Informe-se com o administrador do Wiki, caso você ache que isso está incorreto.
====== Agent-Based Models ======



====An agent-based computational study of wealth distribution in function of resource growth interval using NetLogo====
|R Damaceanu, 2008| Applied Mathematics and Computation 201:371–377| [[http://www.leg.ufpr.br/~pedro/papers/amc/wealth-distribution-resource-growth.pdf|pdf]]|

\\

**Abstract:** We describe an agent-based computational model that simulates the distribution of wealth in three classes: upper, middle and lower. The experimental data show us that: (1) the wealth of economy based on renewable resources is increasing if
the resource growth interval is decreasing with the condition that the other factors remained unchanged; (2) the wealth of
an economy based on renewable resources is higher in comparison with the wealth of an economy based on nonrenewable
resources. This conclusion stresses the fact that the global economy must focus on using renewable resources because this
approach may increase the global wealth.

\\


====Policy analysis from first principles====
|S Moss, 2002| PNAS 99(3)7267–7274| [[http://www.leg.ufpr.br/~pedro/papers/pnas/policy-analysis-moss.pdf|pdf]]|

{{  http://www.leg.ufpr.br/~pedro/figures/moss-sales-grid-size.jpg}}

\\

**Abstract:** The argument of this paper is predicated on the view that social
science should start with observation and the specification of a
problem to be solved. On that basis, the appropriate properties and
conditions of application of relevant tools of analysis should be
defined. Evidence is adduced from data for sales volumes and
values of a disparate range of goods to show that frequency
distributions are commonly fat-tailed. This result implies that **any
stable population distribution will generally have infinite variance
and perhaps undefined mean**. Models with agents that reason
about their behavior and are influenced by, but do not imitate,
other agents known to them will typically generate fat-tailed time
series data. A simulation model of intermediated exchange is
reported that is populated by such agents and yields the same type
of fat-tailed time series and cross-sectional data that is found in
data for fast moving consumer goods and for retail outlets. This
result supports the proposition that **adaptive agent models of
markets with agents that reason and are socially embedded have
the same statistical signatures as real markets**. Whereas this
statistical signature precludes any conventional hypothesis testing
or forecasting, these models do offer unique opportunities for
validation on the basis of domain expertise and qualitative data.
Perhaps the most striking conclusion is that neither current social
theory nor any similar construct will ever support an effective
policy analysis. However, adaptive agent modeling is an effective
substitute when embedded in a wider policy analysis procedure.

\\


====Constructing and Implementing an Agent-Based Model of Residential Segregation through Vector GIS====
|A. T. Crooks, 2008| UCL Working Paper 133| [[http://www.leg.ufpr.br/~pedro/papers/abm-segregation-gis.pdf|pdf]]|


\\

**Abstract: ** In this paper, we present a geographically explicit agent-based model, loosely
coupled with vector GIS, which explicitly captures and uses geometrical data and
socio economic attributes in the simulation process. The ability to represent the
urban environment as a series of points, line and polygons not only allows one to
represent a range of different sized features such as houses or larger areas portrayed
as the urban environment but is a move away from many agent-based models
utilising GIS which are rooted in grid-based structures. We apply this model to the
study of residential segregation, specifically creating a Schelling (1971, 1978) type
of model within a hypothetical cityscape, thus demonstrating how this approach can
be used for linking vector-based GIS and agent-based modelling. A selection of
simulation experiments are presented, highlighting the inner workings of the model
and how aggregate patterns of segregation can emerge from the mild tastes and
preferences of individual agents interacting locally over time. Furthermore, the
paper suggests how this model could be extended and demonstrates the importance
of explicit geographical space in the modelling process.

\\

====Is Religion an Evolutionary Adaptation?====
|J Dow,2008| Journal of Artificial Societies and Social Simulation  11(2)| [[http://jasss.soc.surrey.ac.uk/11/2/2.html|html]]|

\\

** Abstract:**   Religious people talk about things that cannot be seen, stories that cannot be verified, and beings and forces beyond the ordinary. Perhaps their gods are truly at work, or perhaps in human nature there is an impulse to proclaim religious knowledge. If so, it would have to have arisen by natural selection. It is hard to imagine how natural selection could have produced such an impulse. There is a debate among evolutionary scientists about whether or not there is any adaptive advantage to religion at all (Bulbulia 2004a; Atran and Norenzayan 2004). Some believe that it has no adaptive value itself and that it is just a hodge podge of of behaviors that have evolved because they are adaptive in other non-religious contexts. The agent-based simulation described in this article shows that a central unifying feature of religion, **a belief in an unverifiable world, could have evolved along side of verifiable knowledge**. The simulation makes use of an agent-based communication model with two types of information: verifiable information (real information) about a real world and unverifiable information (unreal information) about about an imaginary world. It examines the conditions necessary for the communication of unreal information to have evolved along side the communication of real information. It offers support for the theory that religion is an adaptive complex and it disputes the theory that religion is a byproduct of unrelated adaptive processes.

\\

The existence of gods, spirits, and the like cannot be verified by the senses. A belief in them makes no sense from an common evolutionary point of view. The animal whose conception of the world is out of touch with reality should be eliminated by natural selection. The one whose mental images correspond most closely to the real environment should be one to survive. The primary problem of explaining how religion has evolved through natural selection is the problem of explaining the belief in unreal things.

The agents with whom an agent communicates can be chosen randomly from a uniform distribution of the agents, or they can be chosen from a distribution based on the tendency of another agent to engage in unreal communication.


====Articulating land and water dynamics with urbanization: an attempt to model natural resources management at the urban edge====
|R. Ducrot, C. Le Page, P. Bommel, M. Kuper, 2004| Computers, Environment and Urban Systems| [[http://www.leg.ufpr.br/~pedro/papers/ceus/water-dynamics.pdf|pdf]]|



{{  http://www.leg.ufpr.br/~pedro/figures/water-dynamics.jpg?300}}

**Abstract:** In a rapidly urbanizing world, population densities no longer allow for unlimited access to
safe water. Competition for water, often associated with competition for the access to land,
tends to be exacerbated in peri-urban areas. The objective of this paper is to propose a multiagent
model prototype to represent the relationships between urbanization dynamics and land
and water management in a peri-urban catchment area. A spatially explicit pilot model was
developed using the Cormas platform. This prototype deals with a catchment that is the main
drinking water reservoir and spring of a metropolitan area, and is subjected to high urban
pressure and problems of pollution connected to land use and rain. The combined use of cellular
automata, spatialized passive entities and communicating agents allows the articulation
of the connections between hydrological processes (water cycle, pollution), land-use changes
and urbanization. However, the representation is based on simplified dynamics and further
work is needed to develop a simulation model that could be used as a discussion tool for land
and water management at the urban edge.

\\

This work presents a simple model of agents. There is no real data, unless some temporal events such as crop cycles. It studies more the growth of favelas than the water dynamics per se.


====Simulation of common pool resource field experiments: a behavioral model of collective action====
|Daniel Castilloa, Ali Kerem Saysel, 2005| Ecological Economics|

{{ http://www.leg.ufpr.br/~pedro/figures/laboratory-and-model.jpg}}

\\

**Abstract:** We investigate the decision rules adopted by individuals in local communities, whose livelihoods depend on common pool
resource stocks and who face the cooperation dilemma in their everyday life. For this purpose, field experiments are modeled
and the model structure and output are confronted with experimental data and with the relevant theory of collective action
proposed by Ostrom (1998) [Ostrom, E., 1998. A behavioral approach to the rational choice theory of collective action.
American Political Science Review 92 (1), 1–22.]. The field experiments analyze the cooperative action among coastal
communities in Providence Island (Colombian Caribbean Sea). The simulation model is built according to the principles and
methods of System Dynamics. The model formalizes the feedback causality among reputation, trust and reciprocity Ostrom
(1998). Moreover, based on the payoff structure and treatments used in the experiments, it considers other
behavioral factors such as temptation to free ride, profit maximization, awareness and risk perception of the individuals in
feedback perspective. Model behavior replicates the patterns in the experimental data and is highly sensitive to reciprocity and
free-riding behavior. It reveals path-dependent characteristic to the initial trust of the individuals in the group. The variables and
decision rules built into the model structure provide the basis for a dialogue between the theories of collective action and future
experimental designs to test and improve such theories.

\\



====Land use decisions in developing countries and their representation in multi-agent systems====
|P Schreinemachers and T Berger, 2006| JLUS|

\\

**Abstract:** Recent research on land use and land cover change (LUCC) has put more emphasis on the importance of understanding the decision-making of human actors, especially in developing countries. The quest is now for a new generation of LUCC models with a decision-making component. This paper deals with the question of how to realistically represent decision-making in land use models. Two main agent decision architectures are compared. Heuristic agents take sequential decisions following a pre-defined decision tree, while optimizing agents take simultaneous decisions by solving a mathematical programming model. Optimizing behaviour is often discarded as being unrealistic. Yet the paper shows that optimizing agents do have important advantages for empirical land use modelling and that multi-agent systems (MAS) offer an ideal framework for using the strengths of both agent decision architectures. The use of optimization models is advanced with a novel three-stage decision model of investment, production, and consumption to represent uncertainty in models of land use decision-making.

\\

====The (Ir-)Relevance of the Crop Yield Gap Concept to Food Security in Developing Countries With an Application of Multi-Agent Modeling to Farming Systems in Uganda====
|P. Schreinemachers, 2006| Thesis| [[http://hss.ulb.uni-bonn.de/diss_online/landw_fak/2006/schreinemachers_pepijn/|page]]|

\\

{{  http://www.leg.ufpr.br/~pedro/figures/population-network.jpg?600}}

**Abstract:** This thesis scrutinizes the relationship between the width of the crop yield gap and
farm household food security. Many researchers have argued that an exploitable gap
between average crop yields and the genetic yield potential contributes to food
security and that this potential should therefore be improved. Yet, crop yield gaps in
developing countries are mostly wide, which is prima facie evidence that **factors
other than the yield potential are most constraining**. [...]
Multi-agent systems are used to model the heterogeneity in
socioeconomic and biophysical conditions. The model integrates three components:
(1) whole farm mathematical programming models representing human decision making;
(2) spatial layers of different soil properties representing the physical
landscape; and (3) a biophysical model simulating crop yields and soil property
dynamics. The thesis contributes to methodology in four ways: First, it is shown that
**MAS can be parametrized empirically from farm survey data**. Second, it develops a
non-separable **three-stage decision model of investment, production, and
consumption** to capture economic trade-offs in the allocation of scarce resources
over time. Third, a three-step budgeting system, including an Almost Ideal Demand
System, is used to simulate poverty dynamics. Fourth, coping strategies to food
insecurity are included. [...]
It is shown that the existence of maize yield gaps does not signal inefficiencies but poverty can be
reduced substantially by addressing the underlying constraints such as access to
innovations and credit. Improvements in labor productivity are crucial and are a
much better indicator of development than crop yields and yield gaps. The results
suggest that **a strong focus on crop yields and yield gaps might not only be
inefficient but even counterproductive to development.**

\\

The author uses four programs for building his model: MatLab, STATA, MP-MAS and IBM-OSL (a library).

//There is still no commercially available software for using MAS based on
mathematical programming.//
He used a sampling factor of 0.18, and Monte Carlo techniques to generate agent populations.



====Swarming methods for geospatial reasoning====
|H. V. D. Parunak, S. A. Brueckner, R. Matthews, J. Sauter, 2006| IJGIS| [[http://leg.ufpr.br/~pedro/papers/ijgis/parunak_geospatial_reasoning_06.pdf|pdf]]|

\\

**Abstract:** Geospatial data are often used to **predict or recommend movements of robots**,
people, or animals (‘walkers’). Analysis of such systems can be combinatorially
explosive. Each decision that a walker makes generates a new set of possible
future decisions, and **the tree of possible futures grows exponentially**. Complete
enumeration of alternatives is out of the question. One approach that we have
found promising is to instantiate a large population of **simple computer agents
that explore possible paths** through the landscape. The aggregate behaviour of
this swarm of agents estimates the likely behaviour of the real-world system. This
paper will discuss techniques that we have found useful in swarming geospatial
reasoning, illustrate their behaviour in specific cases, compare them with existing
techniques for path planning, and discuss the application of such systems.

\\

Reasoning about the **movements of
entities constrained by topological or topographical features**.
Instead of reasoning logically about how entities might behave in a geospatial environment,
we create models of the entities, situate them in a model of the environment,
simulate their behaviour, and observe what they do. 

Swarming approach lends itself to implementation on parallel hardware,
while the sequential exploration of a branching trajectory is difficult to parallelize
due to the dependence of later steps on those that have gone before.

Characteristics of swarming:


^ number of walkers                | many agents to represent only one entity of the real world, in order to optimize the action |
^ walker's internal logic          | simple representation. the authors compares it with the BDI representation |
^ stochasticity                    | do not always choose the best option. give a chance to the others |
^ stigmergic information exchange  | **indirect communication**, mediated by a shared environment. A common form of stigmergy is the use by **ants** of chemical markers (**pheromones**) that they deposit and sense. Advantages over message-based interaction: simplicity, scalability, robustness, and the ability to take advantage of environmental noise to support the need for stochrasticity in decision making|


Examples:
  - searching for a minimum path with movement constraints (lots of agents to represent a single entity)
  - surveillance system, where agents have to walk in order to cover the space (one to one representation)
  - path based on environmental variables (direction and gradient)






====Spatial Behavior in Groups: an Agent-Based Approach====
|F. S. Beltran, L. Salas and V. Quera, 2006| JASSS|[[http://leg.ufpr.br/~pedro/papers/jasss/beltran_behavior_groups_06.pdf|pdf]]|
\\

**Abstract:** We present an agent-based model with the aim of studying how macro-level dynamics of spatial distances among interacting individuals in a closed space emerge from micro-level dyadic and local interactions. Our agents moved on a lattice (referred to as a room) using a model implemented in a computer program called P-Space in order to **minimize their dissatisfaction, defined as a function of the discrepancy between the real distance and the ideal, or desired, distance between agents**. Ideal distances evolved in accordance with the agent's personal and social space, which changed throughout the dynamics of the interactions among the agents. In the first set of simulations we studied the effects of the parameters of the function that generated ideal distances, and in a second set we explored how group macro-level behavior depended on model parameters and other variables. We learned that certain parameter values yielded consistent patterns in the agents' personal and social spaces, which in turn led to avoidance and approaching behaviors in the agents. We also found that the spatial behavior of the group of agents as a whole was influenced by the values of the model parameters, as well as by other variables such as the number of agents. Our work demonstrates that the bottom-up approach is a useful way of explaining macro-level spatial behavior. The proposed model is also shown to be a powerful tool for simulating the spatial behavior of groups of interacting individuals.

\\

it cites the example of a party, where players move forming groups. how to describe the behaviour?

move to empty cells on a lattice in accordance with an established rule: each agent could move to a cell in its Moore neighbourhood (defined as a 3-cell by 3-cell square with the agent's current location in the center) where the sum of attitude values was maximized.



====Modelling spatial practices and social representations of space using multi-agent systems====
|J Bonnefoy, C L Page, J Rouchier, F Bousquet, 2000| Applications of Simulation to Social Sciences| [[http://leg.ufpr.br/~pedro/papers/pratices-social-representation.pdf|pdf]]|[[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=13926660404399726831|3 citations in Scholar]]|

{{  http://leg.ufpr.br/~pedro/figures/shepherd-forest.jpg?250}}


\\

**Abstract:** This paper demonstrates that multi-agent systems have the capacity to model a region in all its
complexity. An example is developed to show that these tools are not only capable of spatializing and
distributing the behaviour of individuals, but above all, that they allow individuals to integrate different
perceptions of space as well as the constraints imposed on them by a community. A dialectic is established
between individuals, spaces and society, which is used to simulate a region using clearly defined social
representations and spatial practices, which are suitable for testing our geographical theories and hypotheses.

\\

Each shepherd agent takes his flock to graze in a forest that is
divided into groves, leading to a degradation. The 
forest regrows according to a probability that gives priority to regeneration at
the edges. The shepherd agents move through the space at random, memorize the
areas where they have been and form their representations: the state of the whole forest is
then judged on the basis of these partial perceptions; the more forest spaces they
come across, the more their representation is of an abundant forest and vice versa.
Several strategies modelled:

{{http://leg.ufpr.br/~pedro/figures/shepherd-strategies.jpg  }}

  - **Personal strategy:** graze his flock as soon as he comes across the resource. The forest always disappears.
  - **Personal strategy:** if the forest is degraded, he will not graze small groves. agent becomes "aware" of the finite nature of the resource and the forest does not disappears.
  - **Collective strategy:** an average of the individual thresholds is calculated every 10 time steps, and groves smaller than this threshold are excluded from grazing . More forest is maintained but it is fragmented into small groves. Agents estimate more forest than the total amount in the model.
  - **Arrangement strategy**, he allows his flock to graze the groves that are of the size set collectively plus half the difference between that and his own threshold of perception (giving the agent time to reduce the size of his flock. Results very similar to the "collective" strategy.

With MAS, it is possible to understand how  resources are linked in a
social context, and the implications given some representations and decisions.
The agents decide on the restrictions that they impose on themselves for using an environment, they adapt
individually to these social restrictions and, thus, transform their common environment,
then strengthen or change the social rules depending on their degree of satisfaction.





====A Generative Bottom-Up approach to the understanding of the development of rural societies====
{{  http://leg.ufpr.br/~pedro/figures/crop-and-roads.jpg?550}}
|H. Q. Huang, W. Macmillan, 2005|Agrifood Research Reports|[[http://leg.ufpr.br/~pedro/papers/rural-societies.pdf|pdf]]|
\\

**Abstract:** This study analyses the complexity in determining the physical carrying capacity of a society in a specific environmental
setting, and highlights the necessity for developing an agent-based modelling approach. In the context of this generative
bottom-up approach, this study introduces an artificial agricultural society model developed in terms of the complex
interactions among intelligent agents over space and time. Sample simulations results are presented to show the
emergent macroscopic patterns of agricultural land use and of agent's travel and transport networks around a market,
along the variations of society's demography and agents' trading prices at various time scales. The implications of the
simulated results for policy making are also analysed.

\\

  - CA model.
  - DLA ([[http://en.wikipedia.org/wiki/Diffusion-limited_aggregation|Diffusion limited aggregation]]) algorithm for choosing fields to grow crops
  - there are usable and non-usable lands, that can be imported from a GIS.
  - agents can choose among four types of crop. One of them is essential for surviving, and the others are used for selling. They choose a crop that maximizes the profit.
  - costs to transport as a linear function of the distance to the central cell (the city)
  - buyers search for sellers with lowest bidding price, and a seller look for higher bidding prices. They predict their prices to the next year based in their actual prices and a moving-averaging mechanism
  - agents need to gain money to survive and can choose whether to produce a descendant


The simulated results follow [[http://en.wikipedia.org/wiki/Johann_Heinrich_von_Th%C3%BCnen|Von Thünen]] rings of spatial distribution for crops. 
They also demonstrate that it is almost impossible for an agricultural society to maintain a steady state economic growth unless some labours are released.
When the agricultural society reaches its carrying capacity, it still has momentum for growth and can result in the collapse of the whole society.


====Residential segregation in an all-integrationist world====
{{  http://leg.ufpr.br/~pedro/figures/zhang-final.jpg}}
|J. Zhang, 2004| Journal of Economic Behavior and Organization| [[http://leg.ufpr.br/~pedro/papers/jebo/zhang_residential_segregation_04.pdf|pdf]]|[[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=17070524852839774200|8 citations in Scholar]]|

\\
**Abstract:** This paper presents a variation of the Schelling [J. Math. Sociol. 1 (1971) 143; T.C. Schelling, Micromotives and Macrobehavior, Norton, New York, 1978] model to show that segregation emerges and persists even if every person in the society prefers to live in a half-black, half-white neighborhood. In contrast to Schelling’s inductive approach, we formulate neighborhood transition as a spatial game played on a lattice graph. The model is rigorously analyzed using techniques recently developed in stochastic evolutionary game theory. We derive our primary results mathematically and use agent-based simulations to explore the dynamics of segregation.

\\

One agent per cell. Simple satisfiability definition, composed by two linear functions. Schelling: preferences for like-colour neighbours at the individual level can be amplified into high levels of segregation.

====Language Evolution and Population Dynamics in a System of Two Interacting Species====
|K. Kosmidis, 2005|Physica A|[[http://leg.ufpr.br/~pedro/papers/physica/kosmidis_language_evolution_05.pdf|pdf]]|[[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=1463425639783517065|24 citations in Scholar]]|

\\

**Abstract:**    We use Monte Carlo simulations and assumptions from evolutionary game theory in order
to study the evolution of words and the population dynamics of a system made of two
interacting species which initially speak two different languages. The species are characterized
by their identity, vocabulary, and have different initial fitness, i.e. reproduction capability. We
investigate how different initial fitness affects the vocabulary of the species or the population
dynamics by leading to a permanent populational advantage. We further find that the spatial
distributions of the species may cause the system to exhibit pattern formation or segregation.
We show that an initial fitness advantage, even though very quickly balanced, leads to better
spatial arrangement and enhances survival probabilities of the species. In most cases the
system will arrive at a final state where both languages coexist. However, in cases where one
species greatly outnumbers the other in population and fitness, then only one species survives
with its ‘‘final’’ language having a slightly richer vocabulary than its initial language. Thus, our
results offer an explanation for the existence and origin of synonyms in spoken languages.

\\

Each site of the lattice may be empty or it can be occupied by "individuals" of A or by B. One of the four neighbour sites is randomly selected.
In the case of empty the individual moves to it. If it is occupied then the two communicate. After the motion there is a probability that 
reproduction will take place.

Two interacting species which initially speak different languages. The spatial distributions of the species may cause the system to exihbit pattern
formation or segregation. In the most cases the system will arrive at a final state where both languages coexist. The results offer explanation
for the existence and origin of synonymous in spoken languages.





====Modelling adaptive, spatially aware, and mobile agents: Elk migration in Yellowstone====
|D. A. Bennett and  W. Tang, 2006| IJGIS| [[http://leg.ufpr.br/~pedro/papers/ijgis/bennett_elk_migration_06.pdf|pdf]]|

\\

**Abstract:** The potential utility of agent-based models of adaptive, spatially aware, and
mobile entities in geographic and ecological research is considerable. Developing
this potential, however, presents significant challenges to geographic information
science. Modelling the spatio-temporal behaviour of individuals requires new
representational forms that capture how organisms store and use spatial
information. New procedures must be developed that simulate how individuals
produce bounded knowledge of geographical space through experiential learning,
adapt this knowledge to continually changing environments, and apply it to
spatial decision-making processes. In this paper, we present a framework for the
representation of adaptive, spatially aware, and mobile agents. To provide
context to this research, a multiagent model is constructed to simulate the
migratory behaviour of elk (Cervus elaphus) on Yellowstone’s northern range. In
this simulated environment, intelligent agents learn in ways that enable them to
mimic real-world behaviours and adapt to changing landscapes.

\\

If we can understand how landscape-level changes affect the spatial and temporal behaviour of elk, we can
better manage this regionally important resource. Gaining this understanding, however, through the direct manipulation 
of the landscape is not feasible.

{{  http://leg.ufpr.br/~pedro/figures/stknowledge.jpg}}


An agent must have the following properties:
  * motivation: why move? (why becomes when)
  * decision-making: move to where?


In mechanistic models, it is assumed that agent's spatial knowledge is limited to its perceptual range and that it has limited ability to remember past experiences or predict future states. Another possibility is to have omnipresent that can calculate the best location in the space to stay, and therefore does not need learning.
Common sense and field studies suggest that reality falls somewhere between these two modelling extremes.


Agents have two forms of spatial memory: episodic (previously visited places for short time periods) and reference
(long-term navigational information).
The patch structure of an elk is represented as a bidirected graph that captures the spatial memory.
It is isomorphic to the landscape structure, and edges have weights representing the attractive force between nodes.
Large-scale movement decisions are based on patch-level information, but the elks move in a grid with higher resolution. 
Therefore we have a [[http://leg.ufpr.br/~pedro/figures/elk-multiscale.jpg|multiscale]] representation.

The spatial distribution of resources and threats can be conceptualized as a field of
attractive and repulsive forces that can dynamically affect the
movement pattern of mobile agents. Through repeated interaction with the
environment, elk can learn these patterns and adapt to associated changes.

They implemented the framework in C++ and apparently it does not have any connection with GIS.

The goal is to maximize elk fitness by learning an optimal set of
edge weights through generations, using genetic algorithms and neural networks.
During each generation, simulated elk attempt to migrate from summer to winter, 
to maximize end-of-winter body mass at the population level. The authors 
use intermediate recombination to produce the next generation of solutions.

In the beginning of the simulation, it is assumed that elk have no spatial knowledge,
and agents decide their movement based on the local neighbourhood and in a large-scale target.
Every decision is a random choose based on a set of weights.




====Spatially explicit experiments for the exploration of land-use decision-making dynamics====
|T P Evans, W Sum and H. Kelley, 2006|IJGIS| [[http://leg.ufpr.br/~pedro/papers/ijgis/evans_experiments_land_use_06.pdf|pdf]]| [[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=543152219099749935|1 citation in Scholar]]|

\\

**Abstract:**  We explore the special outcomes of decision-making through two laboratory-based experiments, one with a homogenous (//sic//) 
land suitability surface and another
with a heterogeneous suitability surface. Subjects make resource allocation
decisions on an abstract landscape and are given a monetary incentive to
maximize their revenue during the experiment. These experimental results are
compared with simulation output from an agent-based model run on the same
abstract landscape that uses a utility-maximizing agent. The main findings are:
(1) landscapes produced by subjects result in greater patchiness and more edge
than the utility-maximization agent predicts; (2) there is considerable diversity in
the decisions subjects make despite the relatively simple decision-making context;
and (3) there is greater deviation of subject revenue from the maximum potential
revenue in early rounds of the experiment compared with later rounds,
demonstrating the challenge of making optimal decisions with little historical
context. The findings demonstrate the value of using non-maximizing agents in
agent-based models of land-cover change and the importance of acknowledging
actor heterogeneity in land-change systems.

\\

{{ http://leg.ufpr.br/~pedro/figures/resources-suitability.jpg}}

We use the term 'experiment' to refer to a laboratory experiment where human subjects are faced with
a specific decision-making task rather than to refer to a model run or simulation.

Many land-cover change models employ utility-maximizing agents in their decision-making calculus.
However, if we find evidence that agents do not make utility-maximizing decisions, as has been debated, 
then what are the spatial implications of non-utility-maximizing behaviour?

The players must decide between two resources (named 'B' and 'G') for each cell he owns in each time step, 
and then receives his payoff. The resources have different prices along the time
(one low-ascending and the other high-descending, with equal prices in the middle of the simulation), and
different suitability in the space. The agents have perfect information, despite the players, that do not 
know any value initially. Players can learn and evolve their strategies based on their payoffs, and they 
can see the chooses of the other players in the last round.

One challenge with experimental research is how to assess whether laboratory subjects are representative
of real-world actors. 







====From households to urban structures: space representations as engine of dynamics in multi-agent simulations====
|J. Bonnefoy, 2001| International Colloquium of theoretical and quantitative geography| [[http://leg.ufpr.br/~pedro/papers/households-urban-structures.pdf|pdf]]|

\\

**Abstract:** the major idea of this paper is to simulate the construction of a segregate urban structure by means of spatial representations. To maintain such a project, the model must be quite simple because we wish to simulate and understand processes of production of space, their temporality and their dynamic influence on individual learning of space, rather than to reproduce real situation. Furthermore, this work is included in a research project of UMR 6012, which aims to do a spatial analysis centred on the analysis of processes in a context of weak quantification. We use a multi-agent system in order to implement complex process: a spatial structure at macro level emerging from simple rules established at micro level and a feedback from macro to micro level. A model is constructed as following: simulated households wander through the multi-agent universe to find the best place of residence according to constraints as their incomes or the prestige of places. They learn about this space by practice and, consecutively, construct varied representations that are going to participate in the construction of a collective representation of space. These two types of representation will contribute to help each household agent to find the best residential choice. Then, these choices will modify the characteristics of spaces and the new spatial structures will compel household agents in their practices of space, their individual representations and their choices. This feedback based on spatial representations contributes to a great extent to the construction of segregate urban forms.

{{ http://leg.ufpr.br/~pedro/figures/segregation-prestige-classes.jpg?400}}

\\

We want to eliminate, as much as possible, all spatial neighbourhood rules (8-O **!!!**). 
Household agents wander through the multi-agent universe to find an optimal place of residence according to two constraints: their income and the prestige of places. But, if we want to avoid the same-colour neighbour dynamics, we absolutely need something else: prestige of places and especially the representations that the agents build up, are used as intermediary.

There are three categories of agents: low income (orange), medium income (green) and high income (blue).
Initially every cell is empty and with zero prestige.

The model is a repetition of the following steps:
  - Each migrant agent visits a random cell and increases its prestige according to his income (the grey to white colours in the figure represent the prestige, and indicates that the cell is empty).
  - The agent calculates a prestige based in the neighbourhood, and adds to the prestige he has seen time steps before. It will be the evaluation of the city prestige the agent has, resulting of his own city practices.
  - The other agents increase the prestige of their cells.
  - If the cell has a prestige that fits in with the agent expectations, it can live in the cell (the opposite also may occurs: leave the cell if it does not fit anymore).

It is not the neighbourhood that commands agent's behaviour but the representations they build, following their practices and the representation of their
own income group. And, even if we modify the perception of prestige (for example to do not using the neighbourhood to calculate the prestige increase),
simulations show segregation forms (8-O **I DON'T BELIEVE THAT:** if there is no spatial relationship, we could, for instance, redistribute the cells in a way that there is no segregation.)


====Groups of Agents with a Leader====
|O Gigliotta, O Miglino and D Parisi, 2007| JASSS | [[http://jasss.soc.surrey.ac.uk/10/4/1.html|html]]|

\\

**Abstract:** We describe simulations of groups of agents that have to reach a target in a two dimensional environment, the performance criterion being the time taken by the last agent to reach the target. If the target is within a given distance from the agent, the agent moves towards the target; otherwise it moves randomly. The simulations contrast groups with and without a leader, where a leader is a member of the group which other members of the group follow as it moves through the environment. We investigate three factors that affect group performance: (1) group size; (2) the presence or absence of an individual agent with the ability to detect targets at a greater distance than those 'visible' to its companions; (3) the existence of a communication network among group members. The results show that, **in groups without communication, leaders have a beneficial effect on group performance, especially in large groups and if the individual with better than average sensory capabilities is the leader of the group. However, in situations where group members can communicate, these results are reversed, with leaders being detrimental, rather than beneficial, to group performance.**

\\

Primates [...] often have complex social hierarchies (Kummer 1971). When a group of primates moves together in the environment, 
specific individuals play a more important role than others in the initiation of movement and in the choice of the direction 
(Boinski and Garber 2000). Informal and formal social hierarchies are also a characteristic feature of human groups, 
so much so that there exists an entire academic discipline - political science - dedicated to social power in human societies.

If it can detect the target, it goes directly to the target and ignores the leader. But if it can't, it moves in a 
direction which is halfway between (a) the direction it would have taken if it had chosen its direction randomly 
and (b) the current location of the leader.
The presence of a leader allows the group to explore the environment in a more aggregate and therefore faster way.

an individual with superior sensory capabilities is useful to the group only if it is also the leader of the group. :?::?: the paper does not explain why

Communication:
  - topology independent of spatial location
  - based on some spatial radius


=====TODO=====

====Um Simulador Tutorial Multi-Agente para Treinamento da Alocação de Equipes Policiais====
| Vasconcelos, E. and Furtado, V., 2005| XVIII Encontro Nacional de Inteligência Artificial, 892–901|

====Multi-Agent Modeling and Analysis of the Brazilian Food-Poisoning Scenario====
|V. Mysore and O. Gill and R. S. Daruwala and M. Antoniotti and V. Saraswat and B. Mishra, 2006| Agents 2006| [[http://leg.ufpr.br/~pedro/papers/brazilian-food-poisoning.pdf|pdf]]|

\\

**Abstract:** The multi-agent modeling and analysis of catastrophic events raise many challenging
problems since they involve a large, interacting, mobile population with complex
behaviors. This research aims to address these problems through the analysis of
simulations and to aid planning efforts for future catastrophic events through
parameterized stochastic models covering the health care providers, emergency
responders, and affected population. As a test case, we examine the massive outbreak of
Staphylococcus aureus food poisoning that occurred in Minas Gerais, Brazil, in 1998 to
demonstrate and evaluate our tools and techniques. In this incident, 8,000 people
consumed contaminated food at a priest’s ordination. Of these, 81 were admitted to
intensive care units of 26 local hospitals after a triage, and 16 of them eventually expired.
We capture the dynamics of such an outbreak by using two kinds of abstract agents —
hospital and person, further augmented with information and communication channels.
Hospital locations and current capacities are broadcast by the hospital to its patients and
to persons with a radio and subsequently exchanged between neighboring persons. This
“outbreak” model has been implemented in the Java version of Repast 3.0. Most
attributes are scaled to be in the range of 0 to 1, with most behavior being probabilistic.
We document the relative performance of the different simulations by using a range of
parameter values for communication channels, personalities, and triage policies, to
understand their combined effect on the overall survival rates. We also introduce the
XSSYS trace analysis and model checking tool for answering complex temporal logic
queries over Repast traces. We discuss how such simulation-based analysis can become a
rigorous tool in aiding public health policy planning.

\\

====Modeling civil violence: An agent-based computational approach====
{{ http://leg.ufpr.br/~pedro/figures/ethnic-genocide.jpg}}
|J M Epstein, 2002| National Acad Sciences| [[http://leg.ufpr.br/~pedro/papers/civil-violence.pdf|pdf]]| [[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=12373576830065333028|57 citations in Scholar]]|



\\

**Abstract:** This article presents an agent-based computational model of civil
violence. Two variants of the civil violence model are presented. In
the ﬁrst a central authority seeks to suppress decentralized rebel-
lion. In the second a central authority seeks to suppress communal
violence between two warring ethnic groups.

\\

====Agent-based models as laboratories for spatially explicit planning policies====
|A. Ligmann-Zielinska, P. Jankowski, 2007| [[http://www.envplan.com/abstract.cgi?id=b32088|Environment and Planning B]]| no citations in Scholar|
\\

**Abstract:** Agent-based modeling and simulation (ABMS) has been a part of geospatial sciences for over a decade. Most research activities so far have concentrated on either extending complexity theory to spatially explicit phenomena, or on designing computational models and software tools. Only a few of these activities have focused on using ABMS for spatially explicit modeling of real-world policy scenarios. In this paper we present a realistic application of ABMS to simulating alternative futures for a small community in Washington State, USA. We develop an ABMS assessment benchmark that comprehensively covers diverse aspects of a good operational agent-based model. Using an ABMS software tool—CommunityViz Policy Simulator—we generate future development scenarios in the municipality of Chelan, WA based on the County and the City Comprehensive Growth Plans. Simulation results are compared with Washington State projections for growth-management planning. The indication of the highest probability locations of urban growth in the studied community is crucial for environmental and economic planning and decision-making. Endangered salmon protection and recreational and retirement influxes of people from the Puget Sound metropolitan area have a direct impact on future growth of the region. The bottom-up microsimulation allows for interposition of individual decisions and actions into forecasting option generation. The ‘heterogeneity, adaptability, and tractability’ benchmark is instrumental in evaluating CommunityViz Policy Simulator and outlining possible challenges for future development of applied agent-based models.

\\



====VIR-POX: An Agent-Based Analysis of Smallpox Preparedness and Response Policy====
|Eidelson, B.M. and Lustick, I. 2004| JASSS|
\\
biological models of infectious diseases 

====Model Alignment of Anthrax Attack Simulations====
|Chen, L.-C., Carley, K.M., Fridsma, D., Kaminsky, B. and Yahja, A. 2006|Decision Support Systems|
\\
biological models of infectious diseases 

====Social Interactions, Local Spillovers and Unemployment====
|Topa, G. 2001|Review of Economic Studies|
\\
spatial patterns of unemployment 


====Terrorist Organization Modelling====
|North, M., Macal, C.M. and Vos, J.R. 2004|North American Association for Computational Social and Organizational Science Conference|
\\


====Exploring the geographic consequences of public policies using evolutionary algorithms====
|Bennett, D.A., Xiao, N. and Armstrong, M.P., 2004| Annals of the Association of American Geographers|


\\

====Multi-Agent Modelling and Renewable Resources Issues: The Relevance of Shared Representations for Interacting Agents====
|J Rouchier, F Bousquet, O Barreteau, C L Page, J Bonnefoy| [[http://www.springerlink.com/content/aw1cgfrblva944bh/|LNCS]]|[[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=1446389678645429853|18 citations in Scholar]]|

\\

**Abstract:** The issue that is addressed in this paper concerns the way interacting agents should understand their environment so that a common good used by the whole group would last. We synthesise the results of four models with agents interacting in artificial societies in which they have to share a resource. The four societies were built using multi-agent based simulation models that address issues related to the use of common renewable goods. The resources that are used by the artificial communities of agents are of two types: for some, agents must co-ordinate to exploit the resources; for others, the distribution of goods among agents is directly dependent on the distribution of the agents in space. But that classification cannot necessarily hold: the good use of the resources relies on an even distribution of agents in space, but this can be obtained with individual processes in some cases whereas in others it implies coordination too.

\\

====Endogenizing geopolitical boundaries with agent-based modeling====
|L. Cederman, 2002| PNAS | [[http://leg.ufpr.br/~pedro/papers/pnas/geopolitical-bounderies-abm.pdf|pdf]]| [[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=543634779975860919|23 citations in Scholar]]|

\\

**Abstract:** Agent-based modeling promises to overcome the reification of actors. Whereas this common, but limiting, assumption makes a lot
of sense during periods characterized by stable actor boundaries, other historical junctures, such as the end of the Cold War, exhibit
far-reaching and swift transformations of actors’ spatial and organizational existence. Moreover, because actors cannot be
assumed to remain constant in the long run, analysis of macrohistorical processes virtually always requires ‘‘sociational’’
endogenization. This paper presents a series of computational models, implemented with the software package REPAST, which
trace complex macrohistorical transformations of actors be they hierarchically organized as relational networks or as collections of
symbolic categories. With respect to the former, dynamic networks featuring emergent compound actors with agent compartments
represented in a spatial grid capture organizational domination of the territorial state. In addition, models of ‘‘tagged’’ social processes
allows the analyst to show how democratic states predicate their behavior on categorical traits. Finally, categorical schemata
that select out politically relevant cultural traits in ethnic landscapes formalize a constructivist notion of national identity in
conformance with the qualitative literature on nationalism. This ‘‘finite-agent method’’, representing both states and nations as
higher-level structures superimposed on a lower-level grid of primitive agents or cultural traits, avoids reification of agency.
Furthermore, it opens the door to explicit analysis of entity processes, such as the integration and disintegration of actors as well
as boundary transformations.

\\


====Agent-based modeling: Methods and techniques for simulating human systems====
|E Bonabeau, 2002| PNAS| [[http://leg.ufpr.br/~pedro/papers/pnas/abm-simulating-human-systems.pdf|pdf]]| [[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=17132962871516957337|119 citations in Scholar]]|

{{  http://leg.ufpr.br/~pedro/figures/fire-escape.gif}}

\\

**Abstract:** Agent-based modeling is a powerful simulation modeling technique that has seen a number of applications in the last few years,
including applications to real-world business problems. After the basic principles of agent-based simulation are briefly introduced,
its four areas of application are discussed by using real-world applications: flow simulation, organizational simulation, market
simulation, and diffusion simulation. For each category, one or several business applications are described and analyzed.


\\

====Money-scape: a generic agent-based model of corruption====
{{  http://leg.ufpr.br/~pedro/figures/corruption.jpg}}
|H Situngkir, 2004|Computational Economics, EconWPA| [[http://leg.ufpr.br/~pedro/papers/money-scape.pdf|pdf]]| [[http://scholar.google.com.br/scholar?hl=pt-BR&lr=&cites=12303209651617378524|5 citations in Scholar]]|

\\

**Abstract:** There has been a lot of works on corruption cases. We must think that corruption
is a cultural aspect in a social life. We cannot separate the corruption with the cultural
system where the corrupsion (//sic//) raised. Indonesia has been recorded to be one of the contries (//sic//) of
the worst economic and political system on corruption case. The paper is introducing the usage
of agent based simulation for analysing the corruption specifically in Indonesia as the biggest
corruption level. The model showed is named money-scape, inspiring by the way of corrupt
bureaucracy in Indonesia. The paper showed how the simulation taken by revolutionary 
strategies combating corruption and some paceful (//sic//) and smart strategies generically.

\\

Very badly written.

In the figure, darker players are the poorer, and red cells represent jailed bureaucrats.
Each agent has a constant level of honesty and a risk-aversion level that changes over time.
No interesting conclusions.


====Multi-agent simulation of group foraging in sheep: effects of spatial memory, conspecific attraction and plot size====
|B. Dumont and D. R. C. Hill, 2001| Ecological Modelling| [[http://leg.ufpr.br/~pedro/papers/em/spatial-memory-sheep.pdf|pdf]]|

{{  http://leg.ufpr.br/~pedro/figures/bowl-dumont01.jpg}}

\\

**Abstract:** We describe the modelling of sheep spatial memory at pasture using an individual-based approach. As our
modelling goal requires specification of stochastic and state-dependent random movements and some social aspects,
we used a multi-agent system that can be regarded as a special case of an individual-based model (IBM). We used
a three-phase approach to implement the synchronization kernel since this is particularly well adapted to spatial
resource competition. One of the main differences between this model and most earlier IBMs is that we were able to
use real ﬁeld data from animal experiments for model validation. We thus compared real system behaviour with
model predictions. As the simulation results were consistent with ﬁeld data, we used the model as an extrapolation
tool to investigate conditions that had not been tested, or that are not easily amenable to experimentation. This
enabled us to show that conspecific attraction can have disruptive effects on the searching efficiency of foragers in
habitats, where patches deplete rapidly. We also show that the advantages of a good spatial memory vary according
to the size of the environment to be explored.

\\