Urban Network Analysis

(This comment is in __init__.py in the _01_UrbanNetworkAnalysis directory)

This part mainly covers the four-step transportation planning process: Trip Generation, Trip Distribution, Mode Split, and Trip Assignment

Trip Assignment – Transportation Networks

class PyTrans.UrbanNetworkAnalysis.TransportationNetworks.Link(**kwargs)

Class for handling link object in Transportation Networks

Parameters:

• link_id (int) – identifier of link
• length (float) – length of link
• capacity (float) – capacity of link
• alpha (float) – first BPR function parameter, usually 0.15
• beta (float) – second BPR function parameter, usually 4.0
• from_node (int) – id of origin node of link
• to_node (int) – id of destination node of link
• flow (float) – flow on link
• free_speed (float) – free flow speed of link
• v (float) – speed limit of link
• SO (boolean) – True if objective is to find system optimal solution, False if objective is to find user equilibrium

Variables:

• t0 (float) – link travel time under free flow speed
• time (float) – link travel time based on the BPR function

bpr(alpha=None, beta=None, flow=None)

Method for calculating the BPR function

Parameters:

• alpha (float) – first BPR function parameter, usually 0.15
• beta (float) – second BPR function parameter, usually 4.0
• flow (float) – flow on link

Returns:

link travel time

Return type:

float

get_objective_function()

Method for calculating objective function value

Returns:objective function value Return type:float

get_time()

Method for getting link travel time based on the BPR function

This method is used when setting ‘time’ variable

class PyTrans.UrbanNetworkAnalysis.TransportationNetworks.Network(link_file, trip_file, node_file=None, SO=False)

Class for handling Transportation Networks. This class contains methods to read various TNTP format files from the source and methods of network-wide operations

Parameters:

• link_file (string) – file path of network file, which containing various link information
• trip_file (string) – file path of trip table. An Origin label and then Origin node number, followed by Destination node numders and OD flow
• node_file (string) – file path of node file, which containing coordinates information of nodes
• SO (boolean) – True if objective is to find system optimal solution, False if objective is to find user equilibrium

Variables:

• graph (networkx.DiGrapy) – graph of links with Link object and travel time under the current condition
• origins (list) – list of origin nodes
• od_vols (dictionary) – key: tuple(origin node, destination node), value: traffic flow

all_or_nothing_assignment()

Method for implementing all-or-nothing assignment based on the current graph.

It updates link traffic flow

build_datastructure()

Method for opening .tntp format network information files and preparing variables for the analysis

open_link_file()

Method for opening network file, containing various link information

Returns:

• list – list of Link objects having current link condition
• list – list of Node objects

open_node_file(graph)

Method for opening node file, containing position information of nodes

This method adds ‘pos’ key-value pair in graph variable

open_trip_file(demand_factor=1.0)

Method for opening trip tables containing OD flows of each OD pair

demand_factor float

demand factor

update_linkcost()

Method for updating link travel time.

class PyTrans.UrbanNetworkAnalysis.TransportationNetworks.Node(node_id=0)

Class for handling node object in Transportation Networks

Parameters:node_id (int) – identifier of a node

class PyTrans.UrbanNetworkAnalysis.TransportationNetworks.Visualization

Class for handling visualization effect

reLocateAlink(px1, py1, px2, py2, offset=0.5)

Method for adjusting location of a link

Parameters:

• px1 (float) – x coordinate of a node
• py1 (float) – y coordinate of a node
• px2 (float) – x coordinate of another node
• py2 (float) – y coordinate of another node

Returns:

• fx (float) – new coordinate of px1
• fy (float) – new coordinate of py1
• tx (float) – new coordinate of px2
• ty (float) – new coordinate of py2

reLocateLinks(graph)

Method for modifying links in graph

graph: networkx DiGraph

graph to present

Trip Assignment – Frank-Wolfe algorithm

class PyTrans.UrbanNetworkAnalysis.Frank_Wolfe.Run(link_file, trip_file, node_file, SO)

Class of implementing Frank-Wolfe algorithm for networks privided from Transportation Networks for Research Core Team (https://github.com/bstabler/TransportationNetworks)

Parameters:

• link_file (string) – file path of network file, which containing various link information
• trip_file (string) – file path of trip table. An Origin label and then Origin node number, followed by Destination node numders and OD flow
• node_file (string) – file path of node file, which containing coordinates information of nodes
• SO (boolean) – True if objective is to find system optimal solution, False if objective is to find user equilibrium

Variables:

• graph (networkx DiGraph) – graph of links when completing the algorithm
• network (nested dictionary) – dictionary of links information and history of Frank-Wolfe algorithm implementation by iteration
• fwResult (dictionary) – dictionary of theta (optimal move size) and objective function value over iterations

Example

A Quick example

#Set the paths of Transportation Networks file

>>> directory = ".\Data\TransportationNetworks\SiouxFalls\"
>>> link_file = '{}SiouxFalls_net.tntp'.format(directory)
>>> trip_file = '{}SiouxFalls_trips.tntp'.format(directory)
>>> node_file = '{}SiouxFalls_node.tntp'.format(directory)
>>> SO = False

#Implement Frank-Wolfe algorithm

>>> fw = Run(link_file, trip_file, node_file, SO)
>>> fw.showODFlow()
>>> fw.showODFlowMap()

BPR(t0, xa, ca, alpha, beta)

Method for calculating link travel time based on BPR function

Parameters:

• t0 (float) – link travel time under free flow speed
• xa (float) – traffic link flow
• ca (float) – capacity of link
• alpha (float) – first BPR function parameter, usually 0.15
• beta (float) – second BPR function parameter, usually 4.0

Returns:

ta – link travel time under the current traffic flow

Return type:

float

calculateZ(theta)

Method for calculating objective function value

Parameters:theta (float) – optimal move size Returns:objective function value Return type:float

lineSearch()

Method for estimating theta (optimal move size)

Returns:optimal move size (rate) between 0 and 1 Return type:float

showODFlow()

Method for presenting table of the optimal traffic assignment of the Frank-Wolfe algorithm procedure

showODFlowMap()

Method for presenting the traffic assignment result on a map

Trip Assignment – User Equilibrium & System Optimum

Trip Assignment – Path based assignment & Gradient Programming

Trip Assignment – Dynamic Traffic Assignment