Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Integrating Ride App Vehicles into the SFO Curbside Traffic ModelESRI User Conference
San Diego
July 11, 2019
SFO Overview
Terminal Area
2016 TNC staging lot
Current TNC staging lots
Terminal Area
Terminal Area
International Terminal
Terminal 1
Terminal 2Terminal 3
Central Parking Garage
International Parking Garages
Roadways
International Terminal curbsides
Domestic curbsides
Each roadway has an inner (terminal-adjacent) and outer (island) curbside zone
During curbside study:Upper level: All drop-offs, courtesy shuttles, and ride app pick-upsLower level: All other pick-ups
• VISSIM microsimulation models for each of the four curbside roadways
(domestic and international, upper and lower)
– Created in 2008 by SFO and HNTB
– Previously updated in 2011 for the reopening of Terminal 2
Curbside Modeling
• Terminal 2 reopened (April 2011)
• Passenger traffic grew from 41.0 million annual passengers (MAP) to
53.1 MAP
• Transportation Network Companies (TNCs or ride apps) began
operating (licensed in September 2014)
– From 0 (2011) to 14,000 (2016) daily trips
– As of 2019, over 30,000 daily trips
• Roadway congestion increased
From 2011 to 2016…
• Regulated by the California Public Utilities Commission (CPUC)
• Required to maintain CPUC and SFO permits
• Trip fees assessed using an in-app reporting system
– Developed in-house by SFO and licensed to AAAE
– Geofence-based
Ride Apps at SFO
Ride App Growth at SFO
0%
50%
100%
150%
200%
250%
300%
350%
400%
450%
500%
20
15
-09
20
15
-10
20
15
-11
20
15
-12
20
16
-01
20
16
-02
20
16
-03
20
16
-04
20
16
-05
20
16
-06
20
16
-07
20
16
-08
20
16
-09
20
16
-10
20
16
-11
20
16
-12
20
17
-01
20
17
-02
20
17
-03
20
17
-04
20
17
-05
20
17
-06
20
17
-07
20
17
-08
20
17
-09
20
17
-10
20
17
-11
20
17
-12
20
18
-01
20
18
-02
20
18
-03
20
18
-04
20
18
-05
20
18
-06
20
18
-07
20
18
-08
20
18
-09
20
18
-10
20
18
-11
20
18
-12
20
19
-01
20
19
-02
20
19
-03
20
19
-04
Year over Year Growth Rate
YOY growth, pick-ups per 100 deplanements YOY growth, drop-offs per 100 enplanements
2016-17
• In 2016, the Airport decided to perform a Curbside Congestion Study
– First time integrating ride apps into the model
• Data sources
– Traditional surveys
– Tube counts
– Commercial vehicle transponder data (cordon lines)
– Ride app trip data
Curbside Congestion Study
Trip Fee Reporting
SFO_ID TNC_ID U_ID TXN_TYPE LICENSE_PLATE LON LATRIDE_
COUNTLOCAL_TNC_TIMESTAMP
32513 ENTRY -122.393588 37.614347 0 1/3/2018 0:00:00
32513 DROP-OFF -122.38504 37.6152 1 1/3/2018 0:00:00
32513 ENTRY -122.391947 37.614294 0 1/3/2018 0:00:00
32513 PICK-UP -122.38489 37.61512 1 1/3/2018 0:00:03
32513 EXIT -122.396512 37.6132 1 1/3/2018 0:00:04
32513 EXIT -122.397531 37.615374 1 1/3/2018 0:00:04
32513 EXIT -122.401119 37.625795 1 1/3/2018 0:00:04
32512 PICK-UP -122.3898065 37.6148366 0 1/3/2018 0:00:05
32512 PICK-UP -122.384391 37.6156674 0 1/3/2018 0:00:06
32513 ENTRY -122.40012 37.633279 0 1/3/2018 0:00:06
32512 ENTRY -122.4043913 37.6366346 0 1/3/2018 0:00:07
32513 EXIT -122.377664 37.604319 0 1/3/2018 0:00:07
32513 EXIT -122.3967 37.61463 0 1/3/2018 0:00:07
32513 EXIT -122.399456 37.632922 0 1/3/2018 0:00:07
32513 PICK-UP -122.38734 37.6177 1 1/3/2018 0:00:08
32512 ENTRY -122.3793383 37.6020836 0 1/3/2018 0:00:08
32512 ENTRY -122.3998125 37.6330114 0 1/3/2018 0:00:08
32513 PICK-UP -122.38984 37.61499 1 1/3/2018 0:00:11
32512 EXIT -122.406023 37.6431129 0 1/3/2018 0:00:14
32513 PICK-UP -122.3847 37.61524 1 1/3/2018 0:00:14
32513 ENTRY -122.393257 37.613909 0 1/3/2018 0:00:15
32513 ENTRY -122.386904 37.607229 0 1/3/2018 0:00:15
32513 ENTRY -122.390584 37.609635 1 1/3/2018 0:00:15
32513 ENTRY -122.399887 37.633373 0 1/3/2018 0:00:17
32513 PICK-UP -122.38714 37.61778 1 1/3/2018 0:00:17
Trip Fee Reporting Purple: drop-offsGreen: pick-ups
Curbside Features
GIS Curbside Features, Schematic
Begin T3, 2,170 End of delineators separating traffic
merging from the outer roadway, 2,350
Door 8 and AirTrain bridge, 2,730
Door 9, 2,840
Door 9 crosswalk, 2,850
Door 11 crosswalk, 2,950
Door 11, 2,960
End T3, 3,140
2,150 2,250 2,350 2,450 2,550 2,650 2,750 2,850 2,950 3,050
Terminal 3 Schematic View
ModelBuilder Model
Python Script# Import arcpy module
import arcpy
import sys
# Script arguments
Raw_TNC_data_table = arcpy.GetParameterAsText(0) + "\\page$"
if Raw_TNC_data_table == '#' or Raw_TNC_data_table == "\\page$" or not Raw_TNC_data_table:
sys.exit("No file selected.")
# Local variables:
Layer_Name_or_Table_View = "page$_Layer"
Temp_Zones = "Data.gdb\\Temp_Segments"
Temp_Zones__2_ = Temp_Zones
Zones = "Data.gdb\\Upper_road_sections"
Output_folder = "Distribution\\"
# Process: Make XY Event Layer
print("Making XY layer...")
arcpy.MakeXYEventLayer_management(Raw_TNC_data_table, "lon", "lat", Layer_Name_or_Table_View, "GEOGCS['GCS_WGS_1984',DATUM['D_WGS_1984',SPHEROID['WGS_1984',6378137.0,298.257223563]],PRIMEM['Greenwich',0.0],UNIT['Degre
# Process: Feature To Point
print("Feature to point...")
arcpy.FeatureToPoint_management(Layer_Name_or_Table_View, Temp_Zones, "CENTROID")
# Process: Near
print("Near...")
…
arcpy.Near_analysis(Temp_Zones, "Upper_road_sections", "120 Feet", "NO_LOCATION", "NO_ANGLE", "PLANAR")
# Process: Add Join
#arcpy.AddJoin_management(Temp_Zones, "NEAR_FID", "Zones", "OBJECTID", "KEEP_ALL")
# Process: Table to Table
print("Exporting table...")
arcpy.TableToTable_conversion(Temp_Zones, Output_folder, "Output" + str(time.time()) + ".csv", "", "LOCAL_TNC_TIMEST \"LOCAL_TNC_TIMEST\" true true
# Process: Delete
print("Cleaning up...")
arcpy.Delete_management(Temp_Zones, "")
Heat Mapping
Drop-off Data Analysis
TNC drop-offs versus first
Southwest door
TNC drop-offs versus first Delta
door
TNC drop-offs versus first United
door
0
20
40
60
80
100
120
140
0
50
100
150
200
250
300
0
50
100
150
200
250
Confirms, Confirmed by Firsthand Observations
• Complemented traditional curbside validation process (camera and in-
person observation and discussions with Airport Duty Managers and
Police Service Aides)
• Transaction locations aligned with stakeholders’ impressions of
curbside “hot spots”
• For drop-offs, the transactions patterns were also applied to taxis and
private vehicles
Data Analysis
Procedural Changes
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
21
70
21
90
22
10
22
30
22
50
22
70
22
90
23
10
23
30
23
50
23
70
23
90
24
10
24
30
24
50
24
70
24
90
25
10
25
30
25
50
25
70
25
90
26
10
26
30
26
50
26
70
26
90
27
10
27
30
27
50
27
70
27
90
28
10
28
30
28
50
28
70
28
90
29
10
29
30
29
50
29
70
29
90
30
10
30
30
30
50
30
70
30
90
31
10
31
30
31
50
31
70
31
90
32
10
32
30
32
50
Distance Along Curb
Cumulative Distribution of TNC Pick-Ups
Pick-ups - Sample 1 Pick-ups - Sample 2
Procedural Changes
0
0.2
0.4
0.6
0.8
1
1.2
21
70
21
90
22
10
22
30
22
50
22
70
22
90
23
10
23
30
23
50
23
70
23
90
24
10
24
30
24
50
24
70
24
90
25
10
25
30
25
50
25
70
25
90
26
10
26
30
26
50
26
70
26
90
27
10
27
30
27
50
27
70
27
90
28
10
28
30
28
50
28
70
28
90
29
10
29
30
29
50
29
70
29
90
30
10
30
30
30
50
30
70
30
90
31
10
31
30
31
50
31
70
31
90
32
10
32
30
32
50
0
100
200
300
400
500
600
Terminal 3 TNC Activity, 6:00-9:59, Sample 2
CROSSWALK
DOOR
VERTICAL CIRC
DROP-OFF
PICK-UP
Visualization
Blue: Drop-OffsRed: Pick-Ups
• Using GIS tools to perform analysis of TNC activity allowed the Airport
to update its VISSIM model with unprecedented accuracy for:
– Volume
– Desirability of curb segments
• GIS analysis, and subsequent visualizations, allowed the Airport to
make procedural changes to improve traffic flow
• Changes in pick-up or drop-off policy, or in-app behavior, can be
analyzed immediately
Findings