Browsing by Subject "network growth"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    The Emergence of Hierarchy in Transportation Networks
    (Springer-Verlag, 2005) Yerra, Bhanu; Levinson, David M
    A transportation network is a complex system that exhibits the properties of self-organization and emergence. Previous research in dynamics related to transportation networks focuses on traffic assignment or traffic management. This research concentrates on the dynamics of the orientation of major roads in a network and abstractly models these dynamics to understand the basic properties of transportation networks. A model is developed to capture the dynamics that leads to a hierarchical arrangement of roads for a given network structure and land use distribution. Localized investment rules - revenue produced by traffic on a link is invested for that link's own development - are employed. Under reasonable parameters, these investment rules, coupled with traveler behavior, and underlying network topology result in the emergence of a hierarchical pattern. Hypothetical networks subject to certain conditions are tested with this model to explore the network properties. Though hierarchies seem to be designed by planners and engineers, the results show that they are intrinsic properties of networks. Also, the results show that roads, specific routes with continuous attributes, are emergent properties of transportation networks.
  • Thumbnail Image
    Item
    SONG 2 - Simulator of Network Growth version 2
    (2016-07-13) Xie, Feng; Levinson, David M; dlevinson@umn.edu; Levinson, David M; University of Minnesota Nexus Research Group
    Software used in the project Beyond Business as Usual (MnDOT MN/RC-2006-36) This research, extending the Mn/DOT-funded project If They Come, Will You Build It, assesses the implications of existing trends on future network construction. It compares forecast networks (using models estimated on historical decisions developed with previous research) under alternative budget scenarios (trend, above trend, below trend), with networks constructed according to alternative sets of decision rules developed with Mn/DOT and Metropolitan Council staff. The comparison evaluates alternative futures using a set of performance measures to determine whether the network we would get in the absence of a change in policies (allowing historical policies to go forward) outperforms or underperforms the networks developed by applying suggested decision rules. This evaluation methodology enables new decision rules for network construction (building new links or widening existing links) to be tested. The research suggests a path beyond “business as usual”. This research incorporates results from the different scenarios tested. It presents the processes, approaches and development to encode historical decision rules. After analyzing flowcharts developed from the interviews of staff at different levels of government, if-then rules are generated for each jurisdiction. This research then describes the details and processes necessary to run the network forecasting models with various decision rules. Results for different scenarios are presented including adding additional constraints for the transportation network expansion and calibration process details. A comparison and analysis between scenarios is made in order to provide a final conclusion on what scenario will produce the greatest benefit for the Twin Cities Metropolitan Area Transportation Network.