Work package P1: A multi-agent model for urban residents


Contact information

Principal investigators: Prof. Dr. Günter Groß mail.gif

Staff: M.Sc. Anna Lena Bittner mail.gif

Institution: Institute of Meteorology and Climatology (http://www.muk.uni-hannover.de/), Leibniz Universität Hannover (https://www.uni-hannover.de/)

Goals of WP-P1

The assessment of meteorological aspects of quality of life in towns and cities is determined by means of wind comfort, heat stress, radiation exposure, and air pollution. In order to obtain a general view of the environmental load, the spatial distribution of relevant meteorological parameters are used for a mapping and zoning in urban areas. However, such a conventionally applied approach does not take into account the wide spectrum of urban resident characteristics with respect to socio-economic data, behavior and movement patterns. In this work package, a multi-agent model (MAM) will be developed and adopted to PALM-4U in order to estimate a personal burden for individual urban residents due to local atmospheric conditions. Modules for the thermal and actinic complex (WP-P2 and WP-P3, respectively) will be implemented. Also air pollution will be considered (WP-M7). A multi-agent approach offers the possibility to consider socio-economic data of the urban inhabitants. Typically, in specific urban districts a characteristic composition of residents can be found with predominantly elderly people, young families or business people in the inner city. Depending on these groups of residents, their work, their individual behaviour and their specific ways between home and final destination, temperature, wind and radiation have very different effects on the individuals. By defining comfort measures for individual groups, comfortable areas, heavy loaded places and optimized walking paths can be estimated.


Work program

WP-P1.1: Developing of MAM

In the first step, the existing MAM code must be adapted for the PALM-4U model. This involves the development of a MAM implementation which uses core parts of the existing Lagrangian particle model in PALM. In this way, each Lagrangian particle will represent a single agent that can carry several properties, and which can move according to a pathfinding algorithm. In the second step, the pathfinding algorithm will be developed and adapted to the specific urban environment. Characteristic walking paths between starting point and final destination for different groups of residents have to be defined and tested. The variety of possible decisions on different walkways has to be compared and analyzed with respect to effort and significance. Also sensitivity studies are necessary to estimate the required number of agents to get significant and meaningful results.

WP-P1.2: Tests, coupling and sensitivity studies of MAM

The MAM will be coupled to the assessment models of WP-P2 and WP-P3 in order to cover the wide variety of possible loads on humans with respect to temperature, radiation and air pollution load. The relevance of different socio-economic and individual data with respect to the relative precision of the outcomes has to be estimated in order to find, if necessary, indications to modify and optimize the model.

WP-P1.3: Application of MAM

The model will be applied for specified situations in the democities to identify the strengths and weaknesses in order to complement and correct the model. The system will be tested to find optimized walking paths for groups of residents of different vulnerability. In a comparison between the MAM approach and the classical method the advantages of both approaches are demonstrated. A model description, interpretation aids and presentation of the importance of different input parameters is provided.

Last modified 20 months ago Last modified on Apr 11, 2017 9:08:48 AM

Attachments (1)

Download all attachments as: .zip

                                                                                                                                                                                                                                                                                                                                                                               
  | Impressum | ©Leibniz Universität Hannover |