Predictions help to be prepared for problem situations – such as epidemics, panics or traffic problems. Computer scientists have developed a system that enables computer simulations. Patric Seibel asked about it.
Electromobility, traffic, logistics and the coordination of major events pose major challenges for political decision-makers, authorities, planning and crisis teams and business representatives. With the research project “SmartOpenHamburg”, HAW Hamburg in cooperation with the University of Hamburg wants to create a tool that supports organizational and logistical decision-making processes.
[HAW] – If you look at traffic as a result of means of transport, routes and rules, it becomes clear that it is a highly complex system. Therefore, it is practically impossible to predict from the gut which measures have which effects on traffic. In order to artificially run through various “what-if” scenarios, the project “SmartOpenHamburg – A multi-modal decision support system for the Hamburg metropolitan region” is developing a model that replicates the traffic situation in Hamburg – a digital twin, so to speak.
Mobility in a Smart City
The topic of mobility is part of the so-called smart city. This also includes local public transport, intelligent traffic management and parking space design. According to a study, Hamburg has a lot of catching up to do compared with other major German cities, especially in the area of “intelligent traffic control. The Hanseatic city needs a high-performance transport infrastructure and, associated with this, well-functioning freight, commuter and passenger traffic. In order to consciously control changes in mobility behavior, a holistic strategy must be in place for medium and long-term urban and traffic planning.
The SmartOpenHamburg research project started at the beginning of 2018 as a sub-project of the Hamburg information technology platform ahoi.digital. It provides a simulation tool for “what-if” scenarios and is embedded in the smart mobility area of the smart city architecture of the city of Hamburg. As a joint project of HAW Hamburg and the University of Hamburg, it is led by Prof. Dr. Thomas Clemen from the Department of Computer Science. It is composed of an interdisciplinary team with a geographer and five computer scientists.
Hamburg’s digital traffic twin: implementation and collaboration
The digital city developed in the project so far already consists of the entire road and rail network, as well as buildings and their inhabitants. Each inhabitant of Hamburg thus theoretically has a digital twin that plans the routes they have to take during the day and makes decisions for means of transport. For this purpose, the agents, meaning people, can also switch between different modes of transportation and choose the fastest combinations. In doing so, they are driven by the questions “What do I have to do next? How do I get there? And what routes do I need to take today?”. Agents in the model are autonomous entities with their own decision-making and action capabilities.
In order to manage current and future traffic, the project group first documented and analyzed the status quo after the project started in 2018. Only after this inventory can various “what-if” scenarios be run through. In addition to the creation of the model and the simulation, the project also includes the presentation of the results, which will summarize the findings at the scheduled end of the project in 2020.
Different scenarios will be developed depending on the issue at hand. Currently, the traffic is to be simulated as well as possible. In normal city traffic, this naturally includes traffic jams, construction sites, the HVV service, but also the control during major events or the intelligent steering of cruise ship tourists.
“The choice for or against a means of transport is fundamentally influenced by how usable the transport system is,” says Ulfia Lenfers, the geographer and research assistant in the project. “Personal desires or attitudes – such as getting to your destination as quickly, cheaply, comfortably or ecologically as possible – also play a role. These personal value criteria also depend on factors such as the weather, the time of year and day, age or one’s own fitness.”
For this purpose, Prof. Clemen’s project group has access to data from the “Urban Data Platform” and other open data sources. In addition, the research team has a wealth of expertise ranging from model development to artificial intelligence (AI) and Big Data to decision support. The close cooperation with Hochbahn Hamburg, HVV, the Urban Data Platform, the Senate Chancellery or even the State Office for Roads, Bridges and Waterways (LSBG) provides the basis for this.
Prof. Clemen: “One of the main goals of SmartOpenHamburg is to make Hamburg a functioning city with as little moving and parking individual traffic as possible. With the concepts developed, the individual mobility behavior of each person is to be controlled in such a way that, in total, a better mix of all means of transport is created. The choice of means of transport should be easy, happen individually in real time if possible and according to both personal preferences and external influences such as rain or increased traffic volume.”
From this experimental environment, scientists and experts can work together to run through “what-if” scenarios and analyze them for various issues. The development of scenarios is not limited to topics of traffic planning and management, but can also include green space planning, kindergarten needs and much more.
Authers: Ulfia A. Lenfers, Britta Sowa,
edited by Katharina Jeorgakopulos
ahoi.digital – Alliance of Hamburg Universities for Computer Science
ahoi.digital is a joint initiative of the University of Hamburg, HAW Hamburg, TU Hamburg and HafenCity University Hamburg, as well as the Hamburg Ministry of Science, and sees itself as an overarching computer science platform in Hamburg. The fundamental goal of ahoi.digital is to expand computer science at Hamburg’s state universities in order to strengthen cooperation with Hamburg companies, authorities and institutions.
SmartOpenHamburg is a project of the MARS Group (mars-group.org). The research group has been working on large complex systems for many years. By means of agent-based modeling, pedestrian escape scenarios, illegal logging in tropical rainforests, the spread of influenza in public transport, and questions of savannah ecology (BMBF research project EMSAfrica), among others, have already been dealt with here. Only the knowledge of the causes and effects of a realistically simulated behavior is the basic prerequisite for developing target group-specific strategies and measures.