Connected population synthesis for transportation simulation. (June 2019)

• Record Type: Journal Article
• Authors: Zhang, Danqing; Cao, Junyu; Feygin, Sid; Tang, Dounan; Shen, Zuo-Jun(Max); Pozdnoukhov, Alexei
• Is Part Of: Transportation research. Volume 103(2019)
• Page Start: 1
• Page End: 16
• Publication Date: 2019-06
• Subjects: Population synthesis -- Cellular data -- Bayesian networks -- Structural learning -- Mixed integer programming -- Exponential random graph model -- Agent-based modeling -- Transportation simulation; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2018.12.014
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: We propose a method of producing synthetic connected populations required to advance agent-based exploration of phenomena involving social influence on travel behaviours. The method reproduces socio-economic characteristics for the synthetic households using a Bayesian network model estimated from household survey data. The method detects and assigns the population to the observed communities, using an integer programming approach. The method generates social networks following the Exponential Random Graph Models estimated from the available social network data. A scalable implementation of the framework is developed (with the source code made available), and illustrated experimentally for a region of the San Francisco Bay Area. Abstract: Agent-based modeling in transportation problems requires detailed information on each of the agents that represent the population in the region of a study. To extend the agent-based transportation modeling with social influence, a connected synthetic population with both synthetic features and its social networks need to be simulated. However, either the traditional manually-collected household survey data (ACS) or the recent large-scale passively-collected Call Detail Records (CDR) alone lacks features. This work proposes an algorithmic procedure that makes use of both traditional survey data as well as digital records of networking and human behavior to generate connected synthetic populations. The generated populations coupled with recent advances in graph (social networks) algorithms can be used for testing transportation simulation scenarios with different social factors.