Presentación de PowerPoint

LT1: Development, calibration and validation of
bus-following model to support analyses and
evaluation of alternative BRT strategies under
different scenarios
Luis Antonio Lindau
Paula Manoela dos Santos
MileOUTPUTS
stones
M1
M2
End of car-following literature review
Report describing potential of existing models to represent bus behavior
M3
Specification requirements for data collection equipment
M4
Plan for field data collection
M5
Data collection
M6
Data analyzes
M7
Development of proposed bus-following model
M8
Final report
Transit Leaders Roundtable
MIT, June 2011
Commercial softwares vs. Car-following models
CA
AIMSUN2
AUTOBAHN
CARSIM
CASIMIR
CORSIM
DRACULA
FOSIM
FREEVU
FRESIM
HUTSIM
INTEGRATION
MICSTRAN
MISSION
MITRAM
MITSIM
NETSIM
PADSIM
Paramics
SIGSIM
SISTM
SITRA-B+
SUMO
VISSIM
Gipps
GHR
Helly
Lógica difusa
Psicofísico
Pipes
Pitt
NETSIM Van Aerde
Main findings
Models are of three types:
• Psychophysical (eg. Wiedemann – VISSIM)
• Stimulus-response (eg. GHR)
• Safe following distance (eg. Gipps)
Model
Number of
parameters
Calibration challenge
Gipps
Low
Low
GHR
Low
Medium
VISSIM
High
High
Comparing: GHR and Gipps
Both are fine for decelerating and resuming acceleration (without stopping)...
0,6
0,4
Aceleração (m/s²)
0,2
Veículo 1
Veículo 2
0
0
20
40
60
80
100
120
140
Veículo 3
Veículo 4
-0,2
Veículo 5
Veículo 6
-0,4
-0,6
-0,8
Tempo (s)
Comparing: GHR and Gipps
... but GHR reveled unstable for typical bus stopping maneuvers.
1
0,5
0
Aceleração (m/s²)
0
20
40
60
80
100
120
140
-0,5
Veículo 1
Veículo 2
-1
-1,5
-2
-2,5
Tempo (s)
Comparing: GHR and Gipps
Gipps is fine for long bus convoys and…
0,6
0,4
Aceleração (m/s²)
0,2
10
1
0
0
20
40
60
80
100
-0,2
-0,4
-0,6
-0,8
Tempo (s)
120
140
160
180
200
Comparing: GHR and Gipps
… represents well the operation of successive long bus convoys.
20
10
18
1
16
11
14
Velocidade
12
10
8
6
4
10 & 20
2
1 & 11
0
550
750
950
1150 stop
bus
Posição
1350
1550
What is next: calibration challenge (manual vs. automatic)
Is there a way of obtaining detailed position from image recognition
of films taken from different locations and angles?
Belo Horizonte: Downtown merging of BRT corridors
BRT Antonio
Carlos
17 k pass arriving
downtown in the
morning peak-hour
BRT Cristiano
Machado
Rio de Janeiro: up to 60k pass/h/direction
Até futuro entroncamento
com TransCarioca –
42 mil pax hora sentido
Deodoro até Terminal
Margaridas – menos que
20 mil pax hora sentido
Entre TransCarioca e Final
da Av. Brasil –
50 a 60 mil pass/h/sentido
Entre Final da Av. Brasil e
Final da Fco Bicalho –
45 mil pax hora sentido
Entre Final da Fco Bicalho e
Candelária –
20 a 25 mil pax hora sentido
TransBrasil + TransCarioca = 19 BRT routes
Work in progress: route optimization (PUC Chile) and performance
simulation (EMBARQ Brasil)
Performance simulation: EMBARQ BRT Simulator
Conferences and submitted publications
•
•
•
PEREIRA, B. M. ; LINDAU, L.A. ; CASTILHO, R. A. A importancia de simular
sistemas Bus Rapid Transit. In: XVI Congreso Latinoamericano de Transporte Publico
y Urbano, 2010, Cidade do Mexico. Anais do XVI CLATPU. Buenos Aires: ALAPTU,
2010. v. 1. p. 1-11.
LINDAU, L.A. ; PEREIRA, B. M. ; CASTILHO, R. A. ; DIÓGENES, M. C. ;
HERRERA, J. C. . Impact of design elements on the capacity and speed of Bus Rapid
Transit (BRT): the case of a single lane per direction corridor. In: Thredbo 12
Conference, 2011, Durban. Workshop 2 BUS RAPID TRANSIT. Sydney : Thredbo
Conference, 2011. v. 1. p. 1-12. (Submitted to Research in Transportation
Economics)
LINDAU, L.A. ; PEREIRA, B. M. ; CASTILHO, R. A. ; DIÓGENES, M. C. Impacto
de elementos de projeto no desempenho operacional de sistemas BRT de faixa única
sem ultrapassagem. XXV ANPET, 2011, Belo Horizonte. Panorama Nacional da
Pesquisa em Transportes 2011. (Submitted to Transportes)
LT1: Development, calibration and validation of
bus-following model to support analyses and
evaluation of alternative BRT strategies under
different scenarios
Luis Antonio Lindau
Paula Manoela dos Santos