#521 Complex Adaptive Systems of Systems
Authors: Bonnie Johnson and Alejandro Hernandez
This paper introduces a new class of systems of systems: complex adaptive systems of systems (CASoS). The rise of automation in many systems; and technology ubiquity in general, present some complex operational environments that require highly collaborative CASoS solutions. This paper describes the need for CASoS’s and how they may be applied to address future complex problems. The paper discusses a conceptual framework for CASoS’s including potential characteristics and implementation strategies.
#529 Correlation network with stubborn agents in an opinion dynamic model.
Authors: David Martínez, Carlos Cusgüen and Eduardo Mojica-Nava
In this work, we evaluate the effects stubborn agents have over dynamics behavior in an opinion dynamics model represented through directed and undirected graphs. We propose to use steady state covariance matrix analysis to determine how this kind of agents presence modifies correlated opinions. Steady state covariance can be obtained through the grounded Laplacian matrix (Dirichlet or Gramian Matrix) of a graph and this is an important tool to describe how state trajectories are associated each other in a system. On the other hand, network topology robustness and noise responses can be analyzed by means of Shannon Entropy measure, which gives us some idea about those network configurations that can improve the information transport accomplished in agent connections. In that way, we expect to find a relationship between information and fitness, focussing our analysis in the worth of identify which agents are more correlated and how they are affected by the stubborn ones. Furthermore, we explore scenarios that involve topological network variations, where the behavior is changing aiming determine how consensus deviates according to stubborn agent position into the network. Finally, we make an approximation to explore the relationship between correlation and fitness giving special attributes to those more correlated trajectories in terms of evolution. We inspect the connection between correlation, network topology robustness, and fitness to find a new way to describe dynamic behavior in opinion dynamics field. We expect to use the obtained results in other network fields, and a general description of the relation between fitness and correlation as a future work.
#533 A new scientific collaboration network model
Authors: Ying Fan, Jianlin Zhou and Zengru Di
Scientific collaboration plays an important role in the knowledge production and scientific development. The researchers have constructed several network models of scientific collaboration. In traditional collaboration network, two scientists are linked if they have coauthored one paper. However, this construction of network undervalues the role of the first author. In this paper, we propose a new collaboration network model considering the importance of the partnership between the first author and others. We make an empirical analysis based on the data of American Physical Society (APS).The results show that there are some t differences of statistical properties between the new network and the traditional one. And the node importance is studied on the new network to identify potential researchers.
#537 Revealed Functions of Endogenous Supply and Demand in Online Games
Authors: Kevin Hoefman, Aaron Bramson and Koen Schoors
We are interested in how the quasi-stable price/quantity relationships of goods and services are endogenously generated by imperfectly informed agents acting towards multiple independent goals. To get accurate empirical data we use data collected from a massively multiplayer game called EVE Online in which players engage in activities such as mining, industry, trade, shipping, exploration, politics, and conflict in an open-ended sandbox environment. The prices of goods in this game, from raw ores and blueprints to sophisticated spaceships composed of hundreds of specialized parts, vary endogenously through the market (and other) behaviors of the players. We use this dataset to evaluate the applicability of core principles in microeconomic behavior of consumers and markets.
In one analysis we compare the objective quality (in terms of the performance values) of a large selection of items with the price players are willing to pay for them. Most of these items have over a dozen variations, and from these we can construct an indifference curve for the marginal performance of the items. These we assess the curves to better understand theories regarding luxury goods versus performance goods versus simply scarce goods. We address which is the best explanation for the observed pattern: status, slight edge over other players, investment, excessive wealth, collecting, or something else?
We also analyze whether goods in EVE follow theories of market clearing quantities and prices. To do this we examine both the supply and demand sides for how many goods are put up for sale and are purchased based on the current market prices at that location. There is always a spectrum of prices across regions and time, but we can determine how well the conditions are satisfied locally, and in what scenarios they are not.
#538 Acclimate - a dynamic, agent-based model for economic loss-propagation with flexible network connections
Authors: Sven Willner, Christian Otto, Leonie Wenz, Katja Frieler and Anders Levermann
Risks of extreme weather events like floods, hurricanes, and droughts are likely to increase under global warming. Additionally, world markets are highly interlinked and local economies are extensively relying on global supply and value added chains, which form a complex global network. In that, extreme weather in one place may have repercussions elsewhere. Accordingly, propagation of losses and nonlinear effects play a crucial role for a comprehensive climate risk management and cost estimation.
Here, we propose the agentbased dynamic loss propagation model "acclimate". On the same time scale as the local events the model explores immediate response dynamics as well as the subsequent recovery phase of the economy. It tries to strike a balance between high flexibility of general equilibrium models and high rigidity of inputoutput models by incorporating heterogeneous agents with bounded rationality but a locally optimal, nonadhoc decision rationale: on a daily timescale each agent optimizes its purchase, demand distribution, as well as production and supply distribution to maximize its revenue. Due to the unanticipated nature of the events we assume limited foresight for the agents performing their optimization locally in a spatial as well as temporal sense. Since the time scale is too short to allow for immediate market clearance local price deviations are hereby taken into account. We furthermore allow for dissipation effects using storage capacities and temporary production extension. The network that determines the agent interactions is built up from multiregional inputoutput data. However, we include a rationale for agents to rewire their connection in reaction to current supply or demand deficit. Thereby, the system finds new, possibly more efficient, equilibria on the longrun after transient recovery dynamics has decayed.
Authors: Bonnie Johnson and Alejandro Hernandez
This paper introduces a new class of systems of systems: complex adaptive systems of systems (CASoS). The rise of automation in many systems; and technology ubiquity in general, present some complex operational environments that require highly collaborative CASoS solutions. This paper describes the need for CASoS’s and how they may be applied to address future complex problems. The paper discusses a conceptual framework for CASoS’s including potential characteristics and implementation strategies.
#529 Correlation network with stubborn agents in an opinion dynamic model.
Authors: David Martínez, Carlos Cusgüen and Eduardo Mojica-Nava
In this work, we evaluate the effects stubborn agents have over dynamics behavior in an opinion dynamics model represented through directed and undirected graphs. We propose to use steady state covariance matrix analysis to determine how this kind of agents presence modifies correlated opinions. Steady state covariance can be obtained through the grounded Laplacian matrix (Dirichlet or Gramian Matrix) of a graph and this is an important tool to describe how state trajectories are associated each other in a system. On the other hand, network topology robustness and noise responses can be analyzed by means of Shannon Entropy measure, which gives us some idea about those network configurations that can improve the information transport accomplished in agent connections. In that way, we expect to find a relationship between information and fitness, focussing our analysis in the worth of identify which agents are more correlated and how they are affected by the stubborn ones. Furthermore, we explore scenarios that involve topological network variations, where the behavior is changing aiming determine how consensus deviates according to stubborn agent position into the network. Finally, we make an approximation to explore the relationship between correlation and fitness giving special attributes to those more correlated trajectories in terms of evolution. We inspect the connection between correlation, network topology robustness, and fitness to find a new way to describe dynamic behavior in opinion dynamics field. We expect to use the obtained results in other network fields, and a general description of the relation between fitness and correlation as a future work.
#533 A new scientific collaboration network model
Authors: Ying Fan, Jianlin Zhou and Zengru Di
Scientific collaboration plays an important role in the knowledge production and scientific development. The researchers have constructed several network models of scientific collaboration. In traditional collaboration network, two scientists are linked if they have coauthored one paper. However, this construction of network undervalues the role of the first author. In this paper, we propose a new collaboration network model considering the importance of the partnership between the first author and others. We make an empirical analysis based on the data of American Physical Society (APS).The results show that there are some t differences of statistical properties between the new network and the traditional one. And the node importance is studied on the new network to identify potential researchers.
#537 Revealed Functions of Endogenous Supply and Demand in Online Games
Authors: Kevin Hoefman, Aaron Bramson and Koen Schoors
We are interested in how the quasi-stable price/quantity relationships of goods and services are endogenously generated by imperfectly informed agents acting towards multiple independent goals. To get accurate empirical data we use data collected from a massively multiplayer game called EVE Online in which players engage in activities such as mining, industry, trade, shipping, exploration, politics, and conflict in an open-ended sandbox environment. The prices of goods in this game, from raw ores and blueprints to sophisticated spaceships composed of hundreds of specialized parts, vary endogenously through the market (and other) behaviors of the players. We use this dataset to evaluate the applicability of core principles in microeconomic behavior of consumers and markets.
In one analysis we compare the objective quality (in terms of the performance values) of a large selection of items with the price players are willing to pay for them. Most of these items have over a dozen variations, and from these we can construct an indifference curve for the marginal performance of the items. These we assess the curves to better understand theories regarding luxury goods versus performance goods versus simply scarce goods. We address which is the best explanation for the observed pattern: status, slight edge over other players, investment, excessive wealth, collecting, or something else?
We also analyze whether goods in EVE follow theories of market clearing quantities and prices. To do this we examine both the supply and demand sides for how many goods are put up for sale and are purchased based on the current market prices at that location. There is always a spectrum of prices across regions and time, but we can determine how well the conditions are satisfied locally, and in what scenarios they are not.
#538 Acclimate - a dynamic, agent-based model for economic loss-propagation with flexible network connections
Authors: Sven Willner, Christian Otto, Leonie Wenz, Katja Frieler and Anders Levermann
Risks of extreme weather events like floods, hurricanes, and droughts are likely to increase under global warming. Additionally, world markets are highly interlinked and local economies are extensively relying on global supply and value added chains, which form a complex global network. In that, extreme weather in one place may have repercussions elsewhere. Accordingly, propagation of losses and nonlinear effects play a crucial role for a comprehensive climate risk management and cost estimation.
Here, we propose the agentbased dynamic loss propagation model "acclimate". On the same time scale as the local events the model explores immediate response dynamics as well as the subsequent recovery phase of the economy. It tries to strike a balance between high flexibility of general equilibrium models and high rigidity of inputoutput models by incorporating heterogeneous agents with bounded rationality but a locally optimal, nonadhoc decision rationale: on a daily timescale each agent optimizes its purchase, demand distribution, as well as production and supply distribution to maximize its revenue. Due to the unanticipated nature of the events we assume limited foresight for the agents performing their optimization locally in a spatial as well as temporal sense. Since the time scale is too short to allow for immediate market clearance local price deviations are hereby taken into account. We furthermore allow for dissipation effects using storage capacities and temporary production extension. The network that determines the agent interactions is built up from multiregional inputoutput data. However, we include a rationale for agents to rewire their connection in reaction to current supply or demand deficit. Thereby, the system finds new, possibly more efficient, equilibria on the longrun after transient recovery dynamics has decayed.
Home |
Calls |
Programme |
Registration |
Contact |
Copyright CCS © 2016