An Organic Computing Approach to Resilient Traffic Management

Sommer, Matthias; Tomforde, Sven; Hähner, Jörg

doi:10.1007/978-3-319-25808-9_7

Cited by 24 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We consider traffic control in urban regions as a possible scenario: Assume a control system combining local decisions at each intersection controller with city-based global control strategies (see [6] for a motivating example). Locally this has access to sensor information (e.g., induction loop and camera sensors), to preprocessed data from neighboured intersection controllers, and to established models (e.g., topology and simulation models reflecting the setup of the infrastructure).…”

Section: Introductionmentioning

confidence: 99%

Reinforcement learning as a basis for cross domain fusion of heterogeneous data

Christensen

Tomforde

2022

Informatik Spektrum

Self Cite

View full text Add to dashboard Cite

We propose to establish a research direction based on Reinforcement Learning in the scope of Cross Domain Fusion. More precisely, we combine the algorithmic approach of evolutionary rule-based Reinforcement Learning with the efficiency and performance of Deep Reinforcement Learning, while simultaneously developing a sound mathematical foundation. A possible scenario is traffic control in urban regions.

show abstract

Section: Introductionmentioning

confidence: 99%

Reinforcement learning as a basis for cross domain fusion of heterogeneous data

Christensen

Tomforde

2022

Informatik Spektrum

Self Cite

View full text Add to dashboard Cite

show abstract

“…Future plans include integrating the concept of adaptive-lane reversal in the Organic Traffic-Control (OTC) system-a self-adaptive and self-organised urban-traffic-management system that controls the green duration of traffic lights [20], establishes progressive signal systems [16] ("green waves") and guides road users through the network using variable message signs [21]. Possible challenges include devising a full decentralised extension of flow-lane reversal as well as using the OTC's ability to adapt its control strategy for the improvements of such an extension.…”

Section: Future Workmentioning

confidence: 99%

Adaptive Approaches for Tidal-Flow Lanes in Urban-Road Networks

Striewski

Thomsen

Tomforde

2022

Future Transportation

Self Cite

View full text Add to dashboard Cite

Every year, traffic congestion costs the global economy billions of dollars in lost productivity, particularly in urban areas. Traffic congestion is a complex problem, as traffic conditions may change at any time. Tidal-flow lanes can be utilised as a feasible traffic-congestion-mitigation strategy to balance the fluctuating traffic demands throughout the day. This paper proposes an adaptive-lane-reversal approach for tidal-flow lanes, to decrease the impact of traffic congestion in urban areas. In order to evaluate the adaptive approach under various traffic conditions, several algorithms and parameter sets are examined, using various network models and traffic demands. As a result, the total travel time of the vehicles in the various networks was decreased by up to 81%.

show abstract

“…Simulation methods, such as agent-based or Monte Carlo simulations, can provide high-resolution results that can be used to assess the effectiveness of various improvements under different scenarios, but they are resource-demanding, hard to scale up, and need calibration to reflect the real world, which makes them only suitable for small-scale networks or networks with a limited number of elements and links [30,31]. Logical methods, such as optimization and game theory, and their applications, are mainly used to address intentional attacks or to draft informed development strategies, depending on the informed payoff value; however, the accuracy of the mathematical formulation of the impacts, each strategy's actual cost, and the probability of predicting each side behavior affects their feasibility [32][33][34]. Analytical methods include complex networks theory (CN) and simple performance measurements, such as pace and shortest path length; these are the most often-used methods in resiliency studies since they are the least resource-demanding and provide metrics with a reasonably acceptable level of accuracy [35][36][37][38][39].…”

Section: Resiliency In Transportation Systemsmentioning

confidence: 99%

Resiliency Assessment of Road Networks during Mega Sport Events: The Case of FIFA World Cup Qatar 2022

Serdar

Al-Ghamdi

2021

Sustainability

View full text Add to dashboard Cite

Hosting Mega Sport Events (MSEs) is a formidable expedition that requires enormous investments and that has the potential to reform the nation’s future and create a lasting legacy. However, the increase in environmental concerns is pushing host cities to adopt a compact event approach. Compactness increases the concentration of the load on host cities’ infrastructures, which have to preserve an acceptable level of functionality under any possible disturbance; in other words, they should be resilient. Among these infrastructures, the road network plays the most prominent role in the fans’ experiences and the event’s success. To assess its resilience during MSE, we proposed a multilevel assessment approach that focuses on the network cohesion and critical trips performance under several disturbance scenarios, including natural hazards, intentional attacks, and accidents. The framework was applied to the Doha road network, since Doha will be a host city for the FIFA World Cup in Qatar in 2022, which exhibited a high level of resilience to intentional attacks and accidents scenarios. However, during the natural hazard scenario (flooding), the network experienced severe fragmentation, signaling weak resilience and highlighting the need to improve storm management plans. Future research could investigate the use of weighted graphs to increase the accuracy or incorporate different assessment approaches into the framework.

show abstract

An Organic Computing Approach to Resilient Traffic Management

Cited by 24 publications

References 18 publications

Reinforcement learning as a basis for cross domain fusion of heterogeneous data

Reinforcement learning as a basis for cross domain fusion of heterogeneous data

Adaptive Approaches for Tidal-Flow Lanes in Urban-Road Networks

Resiliency Assessment of Road Networks during Mega Sport Events: The Case of FIFA World Cup Qatar 2022

Contact Info

Product

Resources

About