We investigate the possibility to apply a method of calculus analytics developed for predicting critical transitions in complex systems to social systems modelled with agent-based methods (ABMs). We introduce this method on the example of an equation-based modelled system and subsequently test it—to our knowledge for the first time—on ABMs. Our experiments show that the method may have wide applicability in the analysis of social systems. The method can help to approximate abrupt and thus unpredictable regime shifts, even though it may be constrained by stochastics and require a bit more experimentation in selecting suitable variables for making it work in ABMs.

No abstract

Motorized transport is one of the main contributors to anthropogenic CO 2 emissions, which cause global warming. Other emissions, like nitrogen oxides or carbon monoxide, are detrimental to human health. A prominent way to understand and thus be able to minimize emissions is by using traffic simulations to evaluate different scenarios. In that way, one can find out which policies, technical innovations, or behavioral changes can lead to a decrease in emissions. Since the effect of CO 2 is on a global scale, a macroscopic model is often enough to find reasonable results. However, NO x emissions can also have a direct, local effect. Therefore, it is interesting to investigate these emissions on a mesoscopic scale, to gain insight into the local distribution of this pollutant. In this study, we used a traffic model that, contrary to most other state-of-the-art traffic simulations, does not require an origin-destination matrix as an input, but calculates it from mobility behavior extracted from a survey. We then generated agents with realistic mobility behavior that perform their daily trips and calculate key features like congestion and emissions for every edge of the road network. Our approach has the additional advantage of allowing to investigate technical, juridical, as well as behavioral changes, all within the same framework. It is then possible to identify strategies that minimize NO x emissions caused by private motorized transport. Evaluation showed good agreement with reality in terms of local and temporal resolution. Especially when looking at the sum of emissions, the main feature for evaluating policies, and deviations between our simulation and available statistics were negligible. We found that, from all scenarios we investigated, the ban of old diesel cars is the most promising policy: By replacing all diesel cars built in 2005 or earlier with petrol cars of the same age, NO x emissions could drop by roughly a third.the city of Lyon [6] found that traffic was the main contributor, responsible for over 50% of total NO x emissions. When considering citizen exposure to NO 2 in urban areas, the relative contribution of the road sector is even bigger [5]. This extent of emissions is not only caused by the higher population density of an urban environment, but also by high congestion. Congested roads lead to an increase in traffic emissions and thus health risks for people in these areas [4]. In order to quantify those health risks, emission inventories created by coupled traffic and emissions models are then fed into meteorological and atmospheric chemistry transport models to yield their effect on air quality [7]. Subsequently, human exposure models link the concentration of pollutants with human factors [8].There is not necessarily a linear relation of the concentration to health effects. Thus, together with data on other adverse substances, the health hazard can finally be modeled [9].In order to combat the negative effects of traffic-related emissions, infrastructural and policy changes in a city's ro...

Novel developments in artificial intelligence excel in regard to the abilities of rule-based agent-based models (ABMs), but are still limited in their representation of bounded rationality. The future state maximization (FSX) paradigm presents a promising methodology for describing the intelligent behavior of agents. FSX agents explore their future state space using “walkers” as virtual entities probing for a maximization of possible states. Recent studies have demonstrated the applicability of FSX to modeling the cooperative behavior of individuals. Applied to ABMs, the FSX principle should also represent non-cooperative behavior: for example, in microscopic traffic modeling, there is a need to model agents that do not fully adhere to the traffic rules. To examine non-cooperative behavior arising from FSX, we developed a road section model populated by agent-cars endowed with an augmented FSX decision making algorithm. Simulation experiments were conducted in four scenarios modeling various traffic settings. A sensitivity analysis showed that cooperation among the agents was the result of a balance between exploration and exploitation. We showed that our model reproduced several patterns observed in rule-based traffic models. We also demonstrated that agents acting according to FSX can stop cooperating. We concluded that FSX can be useful for studying irrational behavior in certain traffic settings, and that it is suitable for ABMs in general.

No abstract