1 Presented at 2008 North American Paramics User Group Meeting 1 Evaluation of Incorporating Hybrid...
-
Upload
tyler-riley -
Category
Documents
-
view
217 -
download
1
Transcript of 1 Presented at 2008 North American Paramics User Group Meeting 1 Evaluation of Incorporating Hybrid...
1
1Presented at 2008 North American Paramics User Group Meeting
Evaluation of Incorporating Hybrid Vehicle Use of HOV Lanes
Lianyu ChuCLR Analytics Inc
July 14, 2008 @ NA Paramics User Group Meeting
2
2Presented at 2008 North American Paramics User Group Meeting
Background
• Many states have demonstrated the effectiveness of HOV lanes
• HOV lane operation is criticized – Underutilization of roadway capacity – Limited ability to shift solo drivers to transit and
carpools
• Conversion to other type of operation – Allowing single occupant hybrid vehicles (SOHV)– In southern California
• From buffer separate to continuous access
– HOT
3
3Presented at 2008 North American Paramics User Group Meeting
HOV lanes
• What is HOV lane?• Northern CA
– Continuous access– Operated during peak periods– HOV 2+ or 3+
• Southern CA– Limited access– Operated 24 hours a day– HOV 2+
• HOV lanes in CA– About 30% of the total of US
4
4Presented at 2008 North American Paramics User Group Meeting
Current Hybrid HOV policy in California
• Starting Jan 2005– Stickers will expire in 2010.
• Vehicle models > 45 miles per gallon – Toyota Prius– Hybrid Honda Civic – Honda Insight
• Maximum number of stickers– Originally, 75K,
• reached at Nov 2006
– Then, decided to offer 10K more• Reaches 85K on 2/2/2007
– No more stickers
5
5Presented at 2008 North American Paramics User Group Meeting
Motivation
• Statewide policy– San Francisco Bay Area
• there is substantial reserve capacity on HOV lanes
– Orange County, • HOV lanes have almost reached their nominal capacity of
1,650 vehicles per hour, carrying an average of 1,568 vph in 1998
• Purpose:– Investigate impacts of the policy in OC
• Operational effects • Emission effects
6
6Presented at 2008 North American Paramics User Group Meeting
Study Network
7
7Presented at 2008 North American Paramics User Group Meeting
Socio-economic characteristics &
Hybrid Population Data
PolicyDesign scenario
Estimate hybrid demand
Select study site
Build micro-simulation model in
Paramics
Calibrate micro-simulation model
Run simulation for each scenario
Performance measures
Estimate SOV / HOV demand
Planning model
Compare simulation results
Policy Implications
Methodology
8
8Presented at 2008 North American Paramics User Group Meeting
Model
• All major freeways– I-5, I-405, SR-55, SR-22,
SR-57, and SR-91– Most freeway mainline
have 4-6 lanes– Excluding 6-mile the
southern part of I-5, the section of SR-91 express lane.
• Model summary– 200 HOV lane miles – 800 mainline lane miles
• Total zones: 265• AM Peak hour model
9
9Presented at 2008 North American Paramics User Group Meeting
HOV lane modeling
• Network construction (buffer-separated HOV) – The mixed-flow lanes and HOV lanes were coded as
two separate links wherever required. – Non-buffered sections were coded as a single link
between ingress and egress points.
• Route choice models– Combination of the use of three routing models
• All-or-nothing• Stochastic route choice• Dynamic feedback
10
10Presented at 2008 North American Paramics User Group Meeting
Model calibrationCalibration data preparation
and data analysis
Initial calibration / setting of route choice models
OD demand estimation
Network performance calibration and validation
Final model
Route choice modification
11
11Presented at 2008 North American Paramics User Group Meeting
Calibration data preparation
• Base year: 2005– Both flow and speed data were collected from PeMS
• Flow data:– Gather data from four different years (2002-2005) – Some on-ramp and off-ramp locations
• estimated based on Caltrans census dataset or mainline data
• Speed data– 5-min speed data were collected from freeways for 3
months – Draw the 50th percentile speed contour maps based
on Tuesday to Thursday’s data– Used to identify freeway bottlenecks.
12
12Presented at 2008 North American Paramics User Group Meeting
Demand Estimation• Pattern OD matrix
– OCTAM: planning model of Orange County, CA– Extracted using MMA analysis in TransCAD
• Fine-tune OD matrix using Paramics OD estimator• HOV demand estimation
– A fixed percentage: 21.7%– Based on California DOT’s HOV report and loop detector data– Trial-and-error based on range from 14% to 25%
• Hybrid demand estimation model :– Estimate each zone’s hybrid percentage– Multinomial regression and binomial regression model
• Social-economic data• Survey data
13
13Presented at 2008 North American Paramics User Group Meeting
HOV volume / total volume (based on 2005 D12 HOV report)
Peak period (6:30 - 9:00) Peak hour (6:30-7:30)
Name Freeway PMHOV
volumeTotal
volume PercentageHOV
volumeTotal
volume Percentage
Los Alisos Boulevard I5 NB 18 4365 31995 13.6% 1883 13122 14.3%
Tustin Ranch Road I5SB R28.3 5805 26021 22.3% 2741 10957 25.0%
Main St I5SB 33.1 4807 25021 19.2% 1995 10628 18.8%
Harbor Blvd I5SB 37.4 3754 20951 17.9% 1833 9099 20.1%
Broadway st I5SB 38.7 4304 21368 20.1% 1785 9183 19.4%
Von Karman I 405 NB 7.4 2887 28182 10.2% 1320 11625 11.4%
Ward St I 405 SB 13.2 3846 30094 12.8% 1598 12749 12.5%
Walnut Ave SR 55 SB 14.2 4492 19051 23.6% 1999 8607 23.2%
Warner Ave SR 55 SB R 8.5 5284 23665 22.3% 2345 10310 22.7%
Yorba Linda Blvd SR 57 SB 18.3 3729 16134 23.1% 1571 6785 23.2%
Harbor SR 91 WB 3.3 3377 18018 18.7% 1636 7818 20.9%
• HOV volume / total volume percentage:
– peak period: 10.2% - 23.1%– peak hour: 11.4% - 25%
• Patterns:– HOV lanes attract more carpoolers
since HOV lane is faster – Congested areas: higher percentages
14
14Presented at 2008 North American Paramics User Group Meeting
Regression Model for Share of Hybrid Vehicles
Variable Coefficient t-statistic
Median Household Income /$10k 0.012364 11.25
Average Household Size 0.002286 1.46
Average Workers per Household 0.018347 2.96
Average Workers / Median Household Income -0.06142 -2.90
Constant 0.186429 26.72
Number of observations 62
R-squared 0.9866
Root MSE 0.00248
15
15Presented at 2008 North American Paramics User Group Meeting
Model Calibration/Validations
Criteria and Measures FHWA targets Model Performance
Hourly flow: Model Vs Observed
GEH Statistic – Individual Link FlowsGEH < 5
> 85% of cases 70% of cases
Speed: Model Vs ObservedMatch bottleneck locations To analyst’s satisfaction Matched all major bottleneck
locations (see graphs)
Visual audits: Individual Link Speeds
Visually acceptable Speed-Flow relationship
BottlenecksVisually acceptable Queuing
To analyst’s satisfaction
To analyst’s satisfaction
Satisfied
Satisfied
Data for baseline model calibration- End of 2004 data from PeMS
16
16Presented at 2008 North American Paramics User Group Meeting
5NB HOV Speed Contour: Observed
5NB HOV Speed Contour: Simulated
6:30
:30
6:40
:30
6:50
:30
7:00
:30
7:10
:30
7:20
:30
7:30
:30
864
1105
1305
1387
1652
200
222
2405
2508
2755
281
2979
309
319
350
3738
381
407
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:0
0
6:2
0
6:4
0
7:0
0
7:2
0
7:4
0
8:0
0
8:2
0
8:4
0
9:0
0
9:2
0
9:4
0
10:0
0
79.168(6.91)
80.898(8.64)
82.358(10.1)
83.628(11.37)
85.958(13.7)
87.608(15.35)
89.688(17.43)
91.078(18.82)
92.848(20.59)
94.458(22.2)
95.458(23.2)
96.758(R24.5)
98.058(R25.8)
100.351(28.1)
102.251(30)
102.851(30.6)
103.651(31.4)
104.501(32.25)
105.251(33)
105.851(33.6)
106.851(34.6)
107.851(35.6)
109.251(37)
110.251(38)
111.551(39.3)
112.751(40.5)
111.851(39.6)
112.951(40.7)
5-15 15-25 25-35 35-45 45-55 55-65 65-75
17
17Presented at 2008 North American Paramics User Group Meeting
6:0
0
6:1
5
6:3
0
6:4
5
7:0
0
7:1
5
7:3
0
7:4
5
8:0
0
8:1
5
8:3
0
8:4
5
9:0
0
9:1
5
9:3
0
9:4
5
10
:0040.98
39.6
38.48
37.48
35.6
34.6
33.8
33.2
32.25
31.4
30.6
29.79
27.43
R25.8
R24.5
23.1
22.1
20.59
19.8
17.49
16.3
13.6
11.91
10.4
9.4
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:30
:30
6:40
:30
6:50
:30
7:00
:30
7:10
:30
7:20
:30
4098
405
393
390
380
3631
354
352
346
333
320
306
304
2835
277
2743
2666
2484
235
2305
2275
2059
2023
198
185
1726
163
1503
136
124
107
86
5-15 15-25 25-35 35-45 45-55 55-65 65-75
5SB HOV Speed Contour: Observed
5SB HOV Speed Contour: Simulated
18
18Presented at 2008 North American Paramics User Group Meeting
6:30
:30
6:35
:30
6:40
:30
6:45
:30
6:50
:30
6:55
:30
7:00
:30
7:05
:30
7:10
:30
7:15
:30
7:20
:30
7:25
:30
9
193
386
403
555
574
685
707
773
965
114
1285
1397
1482
1539
1676
1792
1924
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:0
0
6:1
5
6:3
0
6:4
5
7:0
0
7:1
5
7:3
0
7:4
5
8:0
0
8:1
5
8:3
0
8:4
5
9:0
0
9:1
5
9:3
0
9:4
5
10:0
00.7(0.93)
1.7(1.93)
3.08(3.31)
4.78(5.01)
5.51(5.74)
6.84(7.07)
8.17(8.4)
10.25(10.48)
11.37(11.6)
12.93(13.16)
14.31(14.54)
15.16(15.39)
16.53(16.76)
18.42(18.65)
20.1(20.33)
21.68(21.91)
23.39(23.62)
5-15 15-25 25-35 35-45 45-55 55-65 65-75
405NB HOV Speed Contour: Observed
405NB HOV Speed Contour: Simulated
19
19Presented at 2008 North American Paramics User Group Meeting
6:0
0
6:1
5
6:3
0
6:4
5
7:0
0
7:1
5
7:3
0
7:4
5
8:0
0
8:1
5
8:3
0
8:4
5
9:0
0
9:1
5
9:3
0
9:4
5
10:0
024.12
23.19
21.91
20.33
18.65
17.45
15.87
14.54
13.16
11.5
10.28
8.4
7.01
5.69
5.01
3.31
1.93
0.77
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:30
:30
6:40
:30
6:50
:30
7:00
:30
7:10
:30
7:20
:30
7:30
:30
2069
1798
166
1472
125
112
83
701
569
403
288
96
5-15 15-25 25-35 35-45 45-55 55-65 65-75
405SB HOV Speed Contour: Observed
405SB HOV Speed Contour: Simulated
20
20Presented at 2008 North American Paramics User Group Meeting
6:30
:30
6:35
:30
6:40
:30
6:45
:30
6:50
:30
6:55
:30
7:00
:30
7:05
:30
7:10
:30
7:15
:30
7:20
:30
7:25
:30
694
716
785
919
941
100
102
104
105
1174
139
1522
5-15 15-25 25-35 35-45 45-55 55-65 65-75
55NB HOV Speed Contour: Observed
55NB HOV Speed Contour: Simulated
6:00
6:10
6:20
6:30
6:40
6:50
7:00
7:10
7:20
7:30
7:40
7:50
8:00
8:10
8:20
8:30
8:40
8:50
9:00
9:10
9:20
9:30
9:40
9:50
10:0
0R6.94
R7.85
R9.19
R10
10.4
10.84
11.74
12.7
13.9
14.6
15.4
16.2
5-15 15-25 25-35 35-45 45-55 55-65 65-75
21
21Presented at 2008 North American Paramics User Group Meeting
6:3
0:3
0
6:3
5:3
0
6:4
0:3
0
6:4
5:3
0
6:5
0:3
0
6:5
5:3
0
7:0
0:3
0
7:0
5:3
0
7:1
0:3
0
7:1
5:3
0
7:2
0:3
0
7:2
5:3
0
1612
120
1162
1084
919
1351
812
762
703
688
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:00
6:15
6:30
6:45
7:00
7:15
7:30
7:45
8:00
8:15
8:30
8:45
9:00
9:15
9:30
9:45
10:0
016.2
14.4
13.2
10.84
R10
R7.62
5-15 15-25 25-35 35-45 45-55 55-65 65-75
55SB HOV Speed Contour: Observed
55SB HOV Speed Contour: Simulated
22
22Presented at 2008 North American Paramics User Group Meeting
6:30
:30
6:35
:30
6:40
:30
6:45
:30
6:50
:30
6:55
:30
7:00
:30
7:05
:30
7:10
:30
7:15
:30
7:20
:30
7:25
:30
2206
1994
1973
1837
1818
1741
1718
1646
1483
1465
1345
1327
124
1108
5-15 15-25 25-35 35-45 45-55 55-65 65-75
6:00
6:10
6:20
6:30
6:40
6:50
7:00
7:10
7:20
7:30
7:40
7:50
8:00
8:10
8:20
8:30
8:40
8:50
9:00
9:10
9:20
9:30
9:40
9:50
10:00
22.06
19.73
18.85
18.18
17.18
16.11
14.83
14.4
13.45
12.58
11.55
11.08
5-15 15-25 25-35 35-45 45-55 55-65 65-75
57SB HOV Speed Contour: Observed
57SB HOV Speed Contour: Simulated
23
23Presented at 2008 North American Paramics User Group Meeting
Scenario Construction• Base Case:
– No Hybrids allowed on HOV lanes
– before California’s bill AB 2628 passed
• Scenario 1:– 36K Hybrids in CA (Nov 2005)– 3707 of 36K hybrids in OC
• Scenario 2:– 50K Hybrids (April 2006)– 5216 of 50K hybrids in OC
• Scenario 3:– 75K Hybrids (Nov 2006)
• Scenario 4:– 100K Hybrid (near future)
Scenarios SOV HOV Hybrid
Base Case 78.3 21.7 0.00
scenario 1 76.8 21.7 1.6
Scenario 2 76.2 21.7 2.2
Scenario 3 75.1 21.7 3.2
Scenario 4 74.0 21.7 4.3
* Total trips for study network: 238K
DMV data show:
-About 10% of CA hybrids are in OC
24
24Presented at 2008 North American Paramics User Group Meeting
Performance Measures
• Overall network performance measures: – Vehicle Hours Traveled (VHT)– Vehicle Miles Traveled (VMT)
• Corridor performance measures:– Average corridor speed– Speed
• SAFETEA-LU needs HOV lanes to keep speed higher than 45 mph for 90% of the peak periods
– LOS• Caltrans has the authority to remove “individual HOV lanes
or portions of those lanes,” if traffic condition exceeds LOS C.
• Emissions & Fuel consumptions– Comprehensive Modal Emission Model (CMEM) from UC
Riverside– second-by-second emissions
25
25Presented at 2008 North American Paramics User Group Meeting
HOV Segments for further analysis
Freeways DirectionsDistance(miles)
ML sections
HOVsections
I-405NB 22 4 5
SB 22 4 4
I-5NB 34 7 7
SB 34 6 7
SR-55NB 15 4 3
SB 15 3 3
SR-57NB 11 3 3
SB 11 3 4
SR-91EB 8 3 4
WB 8 3 3
-Total 43 HOV sections
-Study period: 5minutes (for 1-hour data)
26
26Presented at 2008 North American Paramics User Group Meeting
Simulation design
• Assuming no HOV lane violations during simulation. • Simulations period
– One peak hour in AM – Simulate for 1 hour and 30 minutes. – The first 30 minutes of simulation: warm-up period
• MOE– Collected using different plug-ins – Only the last one hour of the simulations were analyzed.
• Base scenario– used as reference to evaluate other scenarios.
• Five runs were conducted per scenario– The results from the median run are used for analysis.
27
27Presented at 2008 North American Paramics User Group Meeting
-2
-1.5
-1
-0.5
0
0.5
1
1.5
SC1 SC2 SC3 SC4
Per
cen
tag
e o
f in
crea
se
VMT VHT
VMT BC SC1 SC2 SC3 SC4
BC 0
SC1 0.3 0
SC2 0.5 0.4 0
SC3 0.75 0.6 0.65 0
SC4 0.98 0.9 0.9 0.75 0
Overall Network Performance
VHT BC SC1 SC2 SC3 SC4
BC 0
SC1 0.9 0
SC2 0.975 0.85 0
SC3 0.98 0.9 0.7 0
SC4 0.99 0.95 0.8 0.4 0
Confidence interval for the comparison
28
28Presented at 2008 North American Paramics User Group Meeting
Comparison of HOV & ML (I-5 & I-405)
0
10
20
30
40
50
60
70
BaseCase
Scenario1
Scenario2
Scenario3
Scenario4
Sp
ee
d (
mp
h)
I-405 NB HOV I-405 NB ML
0
10
20
30
40
50
60
BaseCase
Scenario1
Scenario2
Scenario3
Scenario4
Spe
ed (
mph
)
I-405 SB HOV I-405 SB ML
0
10
20
30
40
50
60
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
I-5 NB HOV I-5 NB ML
0
10
20
30
40
50
60
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
I-5 SB HOV I-5 SB ML
29
29Presented at 2008 North American Paramics User Group Meeting
Comparison of HOV & ML (SR-55&57)
0
10
20
30
40
50
60
70
80
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
SR-55 NB HOV
SR-55 NB ML
0
10
20
30
40
50
60
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
SR-55 SB HOV SR-55 SB ML
0
10
20
30
40
50
60
70
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
SR-57 NB HOV SR-57 NB ML
0
10
20
30
40
50
60
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Sp
ee
d (
mp
h)
SR-57 SB HOV SR-57 SB ML
30
30Presented at 2008 North American Paramics User Group Meeting
Speed Distribution: HOV sections
0
10
20
30
40
50
60
70
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Pe
rce
nta
ge
Spd<15 Spd<25 Spd<35 Spd<45 Spd<55 Spd>55
• With more traffic allowed into HOV lanes, HOV lane performance degrades
31
31Presented at 2008 North American Paramics User Group Meeting
Percentage of Sections & Time periods with speed greater than 45 mph
77
75
72
69
66
60 65 70 75 80
Base Case
Scenario 1
Scenario 2
Scenario 3
Scenario 4
Percentage
32
32Presented at 2008 North American Paramics User Group Meeting
LOS Distribution: HOV lanes
0
5
10
15
20
25
30
35
40
Base Case Scenario 1 Scenario 2 Scenario 3 Scenario 4
Pe
rce
nta
ge
LOS A LOS B LOS C LOS D LOS E LOS F
33
33Presented at 2008 North American Paramics User Group Meeting
Percentage of Sections & Time spent less than 26 veh/mile/lane (LOS C)
81
70
66
60
55
0 20 40 60 80 100
Base Case
Scenario 1
Scenario 2
Scenario 3
Scenario 4
34
34Presented at 2008 North American Paramics User Group Meeting
Comparison of Emissions
-6.00
-5.00
-4.00
-3.00
-2.00
-1.00
0.00
Scenario 1 Scenario 2 Scenario 3 Scenario 4
CO2 CO HC NOx Fuel
• Scenario 4 reduces emissions significantly– due to high share of Hybrids
35
35Presented at 2008 North American Paramics User Group Meeting
Policy decision: multi-objective optimization problem
• Maximize the benefits obtained from emission – SC4 is the best
• Meet FHWA’s SAFETEA-LU’s requirement.– Base-year: doesn’t meet– Practical: allow another 10% HOV lanes to be under 45 mph – SC1-3 can meet
• HOV lanes must continue to provide benefits for existing carpoolers. – MOE: percentage of ave. speed difference between HOV &
mixed-flow lanes – Speed difference in Sc1-SC3 are 21%, 16% and 11%. – An intuitive feeling : at least 15% speed difference
• SC1-2 meet
• Scenario 2 appears likely to satisfy the desires of all stakeholders.
36
36Presented at 2008 North American Paramics User Group Meeting
Findings
• With the increase of Hybrids, the operational performance of HOV lanes degrades
• Overall, Scenario 2 performs more effectively (50K hybrid)– HOV lane speed is reduced by less than 5% in scenario 2
• From air quality perspective, scenario 4 (100K hybrids) outperforms.
• Hybrid HOV policy may have impacted sales of Hybrid vehicles– April 2006: 50K– Nov 2006: 75K– Feb 2, 2007: 85K
37
37Presented at 2008 North American Paramics User Group Meeting
Policy Implication
• The policy is successful in reducing emission by allowing hybrid vehicles using HOV lanes.
• The policy is not appropriate to be applied to the HOV lanes that don’t have reserved capacity.
• 50K hybrid vehicle permits (SC2) throughout the state could be the maximum that the Orange County HOV system can take without much degradation.
• The policy could have been implemented more strategically, i.e. applying to selected freeways and/or selected time periods.
38
38Presented at 2008 North American Paramics User Group Meeting
Project team
CCIT, UC BerkeleyLianyu Chu
UC IrvineWill Recker
David BrownstoneTom Golob
K S NesamaniChris Breiland
Western Michigan UniversityJun Seok Oh
39
39Presented at 2008 North American Paramics User Group Meeting
Thank you!
Questions & Comments