Arrow Research search

Author name cluster

Andreas A. Falkner

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
1 author row

Possible papers

2

ECAI Conference 2020 Conference Paper

ASP-Based Signal Plan Adjustments for Traffic Flow Optimization

  • Thomas Eiter
  • Andreas A. Falkner
  • Patrik Schneider
  • Peter Schüller

Worldwide, many cities spend considerable effort to reduce traffic and specifically to avoid traffic congestions. Adaptive traffic control systems serve this purpose by dynamically adjusting traffic signals for optimizing the traffic flow on intersections. Systems such as SCOOT are based on an “intelligent” combination of different traffic optimization strategies. However, they miss the possibility (i) to add and change on-demand rules to implement new optimization strategies, and (ii) to simulate the outcome of new strategies on-the-fly which is similar to the capabilities of microscopic traffic simulation tools such as SUMO. In order to overcome the above limitations, we present a novel approach for calculating signal phase plans (SPPs) used for optimizations in traffic control systems. Our approach is based on Answer Set Programming (ASP) and combines ASP encodings of an abstract mesoscopic flow model and a strategy for generating possible SPPs. Experimental results shows that traffic simulation can be well approximated and that the generated SPPs improve the traffic flow effectively.

ECAI Conference 2020 Conference Paper

Intelligent Recommendation & Decision Technologies for Community-Driven Requirements Engineering

  • Ralph Samer
  • Martin Stettinger
  • Alexander Felfernig
  • Xavier Franch
  • Andreas A. Falkner

Requirements Engineering (RE) represents a critical phase in the management and planning of software projects. One of the main reasons for project failure is missing or incomplete RE. In order to reduce the risk of project failure, there exists a high and urgent demand for applying intelligent technologies in RE. Since the RE process is mainly decision- and community-driven, Recommender Systems are supposed to be applied in this particular context to support stakeholders in decision-making and, hence, to increase the quality of the decisions taken by the stakeholders. This paper introduces a variety of innovative recommendation tools developed within the scope of the European Horizon 2020 research project OPENREQ. Moreover, we give an overview of user studies conducted to evaluate our approaches and present final results of selected studies. The study results indicate that the developed concepts have the potential to significantly improve the quality of requirements definition and requirements prioritization.

v2026.09.13