Development and evaluation of decentralized route choice and activity scheduling strategies for active modes
Achieving a proper spatial distribution of the crowd is one of the major challenges of people that are responsible for providing a safe and enjoyable atmosphere at places where a lot of people come together. My goal is to develop decentralized individual supporting tools that collaboratively improve this crowd distribution. A strong focus will be on activity location choice, which is an important aspect when studying active mode behaviour . Part of the research will be conducted by developing simple agent-based strategies and evaluate system performance in simulations. The other part will focus on real applications.
Numerous centralized route assignment optimization methods have been discussed in literature. To solve problems with respect to computational intensiveness and information load, some researchers proposed distributed algorithms for route assignment, still aiming for system optima. A challenge for active modes is to include activity choice in such algorithms as well. Another great challenge during this research will be to predict and/or enforce certain levels of users’ compliancy with provided advice. Little is known on this topic, especially regarding active modes.
Having a tool that improves the distribution of crowds, large-scale events and crowded urban areas can be made more enjoyable and potentially more safe. Besides this, more practical insights into human compliancy regarding activity recommendation can help us better understand human activity selection behaviour and the impact of distractions and provide us with knowledge how to build more efficient supporting tools in general.
|Start/end date: 1st March 2016 - March 2020|
Daily supervisor: Winnie Daamen
Promotor: Serge Hoogendoorn