Friedrich Krebs

Emergent value orientation in self-organization of an animat

Abstract The genetic algorithms proposed by Holland [Holland, J.H., 1992/1975. Adaptation in Natural and Artificial Systems. MIT Press Cambridge] have previously been used in computer experiments to study knowledge growth in simple artificial animal models (‘animats’) acting in a simulated environment. In the present paper, this work is extended to include the concepts of orientation theory [Bossel, H., 1977. Orientors of nonroutine behavior. In: H. Bossel (Editor), Concepts and Tools of Computer-Assisted Policy Analysis. Birkhauser Verlag, Basel, pp. 227–265]. Orientors are value orientations emerging in the evolutionary adaptation of systems to their specific environments. The animat experiments demonstrate how knowledge growth during the learning phase leads to balanced attention to basic needs, i.e. satisfaction of basic orientors (effectiveness, freedom of action, security, adaptability in addition to existence and coexistence needs) with some variation of emphasis among individuals in a population. The results of the computer experiments suggest that (multidimensional) value orientation is a basic emergent feature of evolutionary adaptation to environments characterized by sparse resources, variety, fluctuation, and change. Pathological behavior and system failure must be expected if there is insufficient attention to any of the basic orientors. Apart from these basic insights, the approach employed here can be applied more generally to comprehensive assessments of system fitness and performance, and in particular to comparative studies of the feasibility and viability of future development paths.

Considering baseline homophily when generating spatial social networks for agent-based modelling

Social networks have become an important part of agent-based models, and their structure may have remarkable impact on simulation results. We propose a simple and efficient but empirically based approach for spatial agent-based models which explicitly takes into account restrictions and opportunities imposed by effects of baseline homophily, i.e. the influence of local socio-demography on the composition of one’s social network. Furthermore, the algorithm considers the probability of links that depends on geographical distance between potential partners.The resulting network reflects social settings and furthermore allows the modeller to influence network properties by adjusting agent type specific parameters. Especially the parameter for distance dependence and the probability of distant links allow for control of clustering and agent type distribution of personal networks.

On the evolution of social groups during coffee breaks

This paper focuses on the analysis of group evolution events in networks of face-to-face proximity. First, we analyze statistical properties of group evolution, e.g., individual activity and typical group sizes. Furthermore, we define a set of specific group evolution events. We analyze these using real-world data collected at the LWA 2010 conference using the Conferator system, and discuss patterns according to different phases of the conference.

Dynamics of Task Oriented Agent Behaviour in Multiple Layer Social Networks

In numerous real-world situations, people are confronted with tasks that they are unable to fulfil alone. Often, such tasks are characterised by the necessity to include a number of different expertises to their accomplishment. Consequently, people organise themselves into networks aimed at the completion of some specific task. Examples of such situations are to be found in virtually any domain, such as science, economy, or in the context of managing and maintaining natural resources.

A Spatially Explicit Agent-Based Model of the Diffusion of Green Electricity: Model Setup and Retrodictive Validation.

The purpose of this chapter is to propose and illustrate a validation procedure for a spatially explicit ABM of the diffusion of green electricity tariffs in the German electricity market. We focus on two notions of model validity: We report on structural validity by describing the model setup and its empirical and theoretical grounding. Then we challenge simulation results with a rich spatially explicit historical customer data set thus focusing on retrodictive validity. In particular the latter validation exercise can be prototypic for the class of spatially explicit diffusion ABMs in data-rich domains because it systematically scrutinises validity on different levels of agent aggregation.

An Empirically Grounded Model of Green Electricity Adoption in Germany: Calibration, Validation and Insights into Patterns of Diffusion

Spatially explicit agent-based models (ABM) of innovation diffusion have experienced growing attention over the last few years. The ABM presented in this paper investigates the adoption of green electricity tariffs by German households. The model represents empirically characterised household types as agent types which differ in their decision preferences regarding green electricity and other psychological properties. Agent populations are initialised based on spatially explicit socio demographic data describing the sociological lifestyles found in Germany. For model calibration and validation we use historical data on the German green electricity market including a rich dataset of spatially explicit customer data of one of the major providers of green electricity. In order to assess the similarity of the simulation results to historical observations we introduce two validation measures which capture different aspects of the green electricity diffusion. One measure is based on the residuals of spatially-aggregated time series of model indicators and the other measure considers a temporally aggregated but spatially disaggregated indicator of spatial spread. Finally, we demonstrate the descriptive richness of the model by investigating simulation outputs of the calibrated model in more detail. In particular, the results provide insights into the dynamics of the spatial and lifestyle heterogeneity “underneath†the diffusion curve of green electricity in Germany.

Formation and temporal evolution of social groups during coffee breaks

Group formation and evolution are prominent topics in social contexts. This paper focuses on the analysis of group evolution events in networks of face-to-face proximity. We first analyze statistical properties of group evolution, e.g., individual activity and typical group sizes. After that, we define a set of specific group evolution events. These are analyzed in the context of an academic conference, where we provide different patterns according to phases of the conference. Specifically, we investigate group formation and evolution using real-world data collected at the LWA 2010 conference utilizing the Conferator system, and discuss patterns according to different phases of the conference.