highlights of the new brt planning guide
TRANSCRIPT
What’s new in the BRT Planning Guide, 2nd
Edi i ?Edition?
Projected release date: Fall, 2015j ,
Dr. Walter Hook, President, BRT Pl i I l LLCBRT Planning, Intl. [email protected](former CEO, ITDP)
BRT Planning International, LLC
Main ChangesMain Changes
• Chapter 2: Why BRT? (Comparing transit modes) p y ( p g )revised with up to date data, better related to TCQSM 3rd Edition
• Entirely new Service Planning Chapter (6) the• Entirely new Service Planning Chapter (6): the black box explained
• Best practice in Institutional Structures andBest practice in Institutional Structures and Business Planning defined
• More guidance on transit industry transition h d l imethodologies
• Entirely new Infrastructure Design chapter with construction details more station design detailconstruction details, more station design detail
Better correlated to the BRT StandardBetter correlated to the BRT Standard
Ch t 2 R i iChapter 2 Revisions:
Looking for additional data!Feedback welcome!
Comparative Cost Data from about 100 systems:
Developing Developed CountryAverage Cost per Kilometer by Mode
p g p yMode Cost/Km (US 2013) Cost/Km (US 2013)Bus Rapid Transit 9,770,287$ 9,183,885$ Light Rail Transit n a 40 425 978$Light Rail Transit n.a. 40,425,978$ Heavy Rail Transit 92,106,918$ 384,891,792$
Country City Project Project Total Cost (2013 USD)
Length (km) Cost/Km Quality (BRT
Classification)
Brazil Curitiba BRT "Linha Verde" 241,542,000$ 33.8 7,146,213$ GoldBrazil Rio de Janeiro TransOeste 838 258 000$ 54 9 15 268 816$ Gold
Developing Countries
Bus Rapid Transit ProjectsCapital Cost data compiled from 43 BRT systems. Brazil Rio de Janeiro TransOeste 838,258,000$ 54.9 15,268,816$ Gold
Brazil Rio de Janeiro TransCarioca 573,942,000$ 39 14,716,462$ GoldBrazil Belo Horizonte Antônio Carlos-Pedro 1 361,870,000$ 14.7 24,617,007$ [likely Gold]Brazil Belo Horizonte Cristiano Machado 28,284,000$ 7 4,040,571$ GoldBrazil Average 13,157,814$ GoldChina Beijing BRT Line 1 84,058,342$ 79 1,064,030$ BronzeChina Lanzhou Lanzhou BRT 38,340,000$ 9 4,260,000$ SilverChina Changzhou Changzhou BRT line 1 46,123,522$ 24.5 1,882,593$ BronzeChina Guangzhou Guangzhou BRT 110,270,200$ 22.9 4,815,293$ GoldChina Average 3,005,479$ SilverColombia Bogotá TransMilenio Phase 1 761,560,732$ 41 18,574,652$ Gold
y
Conclusions:
No cost gColombia Bogotá TransMilenio Phase 2 1,387,547,763$ 42 33,036,852$ GoldColombia Bogotá TransMilenio Phase 3 909,542,468$ 37 24,582,229$ GoldColombia Barranquilla Transmetro 270,135,988$ 14 19,295,428$ SilverColombia Cali Mio 835,293,533$ 49 17,046,807$ SilverColombia Cartagena Transcaribe 551,884,038$ 13 42,452,618$ Colombia Pereira Megabús 143,811,800$ 27 5,326,363$ SilverColombia Bucaramanga Metrolínea 345,876,940$ 50 6,917,539$ Colombia Average 20,904,061$ GoldIndia Indore Indore iBus BRT 54,125,347$ 11 4,920,486$ [likely Bronze]India Ahmedabad Janmarg BRT Phase 1 + 2 264,313,320$ 88 3,003,560$ SilverI di D lhi D lhi Hi h C it B S t (HCBS) 23 458 613$ 5 8 4 044 588$ B i BRT
No cost difference between developed and
India Delhi Delhi High Capacity Bus System (HCBS) 23,458,613$ 5.8 4,044,588$ Basic BRTIndia Surat Surat BRTS 137,081,466$ 11 12,461,951$ BronzeIndia Pimpri Chinchwad Primpri Chinchwad BRTS 246,719,614$ 44.775 5,510,209$ [likely Bronze]India Average 5,988,159$ BronzeIndonesia Jakarta Transjakarta - Line 12 34,310,346$ 23.8 1,441,611$ Basic BRTIndonesia Jakarta Transjakarta - Line 11 37,028,244$ 11.4 3,248,092$ Basic BRTIndonesia Jakarta Transjakarta - Line 2 & 3 81,378,081$ 14 5,812,720$ BronzeIndonesia Avg 3,500,808$ BasicMexico Monterrey Ecovía Line 1 128,230,227$ 30 4,274,341$ SilverMexico Puebla RUTA, Line 1 123,998,824$ 19 6,526,254$ BronzeMexico Puebla RUTA Line 2 248 665 446$ 20 12 433 272$
developing countries.
No cost Mexico Puebla RUTA, Line 2 248,665,446$ 20 12,433,272$ Mexico Chihuahua Vivebús 77,138,153$ 20 3,856,908$ [likely Silver]Mexico Estado de Mexico Mexíbus Línea 1 - Cd Azteca - Tecamac 125,791,216$ 16.3 7,717,253$ SilverMexico Estado de Mexico Mexíbus Línea 3 Chimalhuacán - Pantitlá 134,024,021$ 14.75 9,086,374$ SilverMexico Mexico City Metrobús Líneas 1-4 578,173,869$ 93 6,216,923$ SilverMexico Mexico City Metrobús Línea 5 63,523,884$ 10 6,352,388$ SilverMexico León Optibús Etapa 1 66,957,736$ 25 2,678,309$ [likey Silver]Mexico Average 6,571,335.86$ SilverSouth Africa Johannesburg Rea Vaya 1a 311,634,023$ 30 10,387,801$ Silver (IA) South Africa Johannesburg Rea Vaya Phase 1b 234,725,000$ 18 13,040,278$ Bronze (IB)South Africa Tshwane A Re YengPhase IA 96,840,800$ 7 13,834,400$ Unknown
No cost difference between Bronze and Silver
South Africa Cape Town MyCiTiPhase IA as of 2010 404,514,085$ 17 23,794,946$ BronzeSouth Africa Average 15,264,356$ Bronze
France Paris TVM Rungis - Croix de Berny RER 107,767,000$ 22 4,898,500$ SilverFrance Rouen TEOR (Phase 1) 205,592,000$ 38 5,410,316$ SilverFrance Average 5,154,408$ SilverUSA Cleveland HealthLine 207,680,000$ 11 18,880,000$ SilverUSA Eugene Franklin Corridor (Green Line) 26,567,460$ 6 4,427,910$ BronzeUSA Los Angeles Orange Line (Original) 375,640,000$ 23 16,332,174$ BronzeUSA Average 13,213,361$ Bronze
9 770 287$
Developed Country Standard. Gold Standard Cost is double
Developing Country Avg 9,770,287$ Developed Country Avg 9,183,885$ Global Average 10,654,383$ Gold Average 16,310,899$ Gold Silver Average 8,725,432$ SilverBronze Average 8,706,686$ Bronze
LRT Data from 13 LRT systems, all in y ,developing world,
Country City Project Project Total Cost (2013 USD)
Length (km) Cost/km Quality (BRT
Classification)
Light Rail Transit Projects
y y j Cost (2013 USD) (km) Classification)
England London Docklands $1,731,600,000 39 $44,400,000 France Besançon Line 1 Tramway 289,175,000$ 15 19,278,333$ France Dijon Line 1+2 Tramway 506,060,000$ 19 26,634,737$ France Le Havre Line 1+2 Tramway 541,490,000$ 13 41,653,077$ France Reims Line 1 Tramway 486,070,000$ 11 44,188,182$ y , ,$ , ,$France Lyon Line 4 Tramway 310,360,000$ 16 19,397,500$ European Average 32,591,971$ USA Charlotte LYNX Blue Line 503,130,000$ 16 31,445,625$ Silver USA Minneapolis METRO Blue Line 902,914,600$ 20 45,145,730$ USA Denver Denver Central Valley Corridor 171,413,006$ 5.3 32,342,077$ BronzeUSA Denver Denver - South West Corridor 225,019,591$ 8.7 25,864,321$ BronzeUSA Portland Portland Blue 1,764,113,368$ 33 53,457,981$ Silver USA Phoenix Phoenix 1,417,894,781$ 20 70,894,739$ BronzeUSA Pittsburgh Pittsburgh LRT 1,001,978,081$ 26.2 38,243,438$ BronzeUSA Average 42,484,844$ Average 40,425,977.76$
Heavy Rail Transit from 27 systems:Heavy Rail Transit from 27 systems: Country City Project Project Total
Cost (2013 USD) Length
(km) Cost/km
Heavy Rail Transit Projects
Brazil Rio de Janeiro Metro Line 4 3,824,000,000$ 16 $ 239,000,000 Brazil Sao Paulo (Line 4) Heavy Rail Transit 2,842,000,000.00$ 14 $ 203,000,000 China Lanzhou Lanzhou Metro Line 1 3,168,000,000$ 34 93,176,471$ China Guangzhou Guangzhou Metro Line 1 1,623,000,000$ 18.5 87,729,730$ China Shenzhen Shenzhen Metro Line3 1,855,408,197$ 33 56,224,491$ China Guangzhou Guangzhou Metro Line 2 1,449,180,328$ 18.284 79,259,480$ China Guangzhou Guangzhou Metro Line 3 2,458,403,066$ 36 68,288,974$ China Shanghai Metro Line 2 1,540,983,607$ 19 81,104,400$
Developing Country
HRT cost on average 4 times more in the g
China Beijing Metro Line 4 2,573,770,492$ 29 88,750,707$ Colombia Medellín Tranvía de ayacucho 324,599,000$ 4 81,149,750$ Colombia Bogotá Metro de Bogotá 3,450,000,000$ 35 98,571,429$ India Delhi Delhi Metro Phase 1 + 2 7,310,987,706$ 167.3 43,699,867$ India Mumbai Mumbai Metro Line 1 811,107,286$ 11.4 71,149,762$ India Hyderabad Hyderabad Metro Phase I 3,900,000,000$ 72 54,166,667$ India Bangalore Bangalore Namma Metro Phase 1 4,427,089,468$ 42.3 104,659,325$ India Kochi Kochi Metro Phase 1 934,630,895$ 25.612 36,491,914$ Indonesia Jakarta MRT 1,539,009,855$ 14 109,929,275$
$ $
more in the developed world
Mexico Mexico City Metro Línea 12 Extension 621,687,640$ 4 155,421,910$ Mexico Mexico City Línea 12 Metro Ciudad de México 2,167,883,661$ 25 86,715,346$ Mexico Zona Metropolitana/Valle d Suburban Rail Line 1 2,109,555,525$ 27 78,131,686$ Mexico Monterrey Línea 3 Tren subterráneo de Monterrey 438,554,217$ 7.5 58,473,896$ South Africa Johannesburg Gautrain 4,100,568,824$ 80 51,257,110$
France Paris Grand Paris Metro expansion $29,500,000,000 200 $147,500,000 UK London Jubilee Line Extension $5,476,800,000 16 $342,300,000 USA Washington, DC Silver Line 3,140,700,000$ 19 $ 165,300,000 USA L A l R d/P l Li 7 167 500 000$ 25 $ 286 700 000
Developed Country
USA Los Angeles Red/Purple Lines 7,167,500,000$ 25 $ 286,700,000 USA New York 2nd Avenue Subway 17,000,000,000$ 17.3 982,658,960$ Average 126,307,609$ Developing Country Avg 92,106,918$ Developed Country Avg 384,891,792$
ConclusionsConclusions
• Main difference between BRT & LRT is cost ofMain difference between BRT & LRT is cost of the rails and electric catenary, cost of the vehicles and cost of a depot near the tracksvehicles, and cost of a depot near the tracks.
• Need better comparative data on operating costs hard to collectcosts, hard to collect.
Main drivers of capacity differencesMain drivers of capacity differences
• Single lane BRT and LRT have very similar capacity. g y p y• LRT has bigger vehicles with more doors where passengers can board simultaneously (minimizing
i bl d ll ti ) BUTvariable dwell time) BUT…• LRT has lower maximum frequency.• Bottleneck for LRT is Usually block length (sets vehicle• Bottleneck for LRT is Usually block length (sets vehicle length, usually under 61 meters) and headways set by traffic signals, usually >90 sec. (can only handle 1 train per signal phase, and in real world 1 train per 2 signal phase)
• Bottleneck for BRT is the station• Bottleneck for BRT is the station
IF BRT has passing lanes and express services…
• BRT can handle much higher frequencies• BRT can simulate the long LRT vehicles by having many buses boarding and alighting at multiple sub‐stations.
• BRT has no constraint at the traffic signal as many• BRT has no constraint at the traffic signal, as many buses can pass through a single signal phase.
• Saturation of the critical station can be reduced by h i id d d d b li it d thaving some corridor demand served by limited stop services that bypass saturated stations.
• These possibilities can be measured (formulas Chapter p ( p7) in the BRT Planning Guide, but are not acknowledged by TCQSM 3rd Edition: The major distinction
Boarding time per door is virtually d l f didentical for BRT, LRT, and HRT
Variable Dwell TimesMode Seconds per doorHRT & LRT Alighting at level 1 39 ‐ 2 0HRT & LRT Alighting at level 1.39 2.0HRT & LRT Alighting with Steps 3.36 ‐ 3.97HRT & LRT Boarding at level 1 11 2 61HRT & LRT Boarding at level 1.11 ‐ 2.61HRT & LRT Boarding with Steps 2.91 ‐ 4.21BRT B di T Mil i 1 2BRT Boarding TransMilenio 1.2Standard at ‐ level BRT boarding 1.6
Theoretical and Observed CapacityTheoretical and Observed Capacity
V hi l it L d F t F C itTheoretical capacities of different rapid transit alternatives
Vehicle capacity Load Factor Frequency CapacityHRT 8 car single track, best imaginable signaling system 1408 0.85 30 35904HRT 8 car double track 1408 0.85 60 71808LRT 8 module module, no turning restrictions, 2 minute signal 632 0.85 15 8058LRT 8 module, no turns allowed, 90 second signal* 632 0.85 20 10744LRT 8 module double track 632 0.85 40 21488BRT Largest Bi‐articulated 220 0.85 60 11220BRT w/ Passing Lanes 220 0.85 193 36000BRT w/ Passing lanes & limiteds bypassing bottleneck station 220 0.85 241 45000
*TCQSM 3rd Edition p 8‐87 provides 20 as the number of trains that can be processed at grade with a 90 second signal. They reach a capacity of 12,000 pphpd by assuming trains with larger capacity than is commercially available or operable in most on‐street contexts.
Observed Capacities: no street level LRT with capacity higher than 6000
P h k di i b d BRT LRT HRT P h k di i b d BRT LRT HRTCorridor Type PPHPD Level Tracks/ Lanes Source
Bogota BRT 37,700 Surface 2 #Guangzhou BRT 27,400 Surface 2 #Istanbul BRT 18,900 Highway 1 #Lima BRT 13,950 Highway 2 #C li BRT 11 100 S f 2 #
Passengers per hour per peak direction observed: BRT, LRT, HRT
BRTCorridor Type PPHPD Level Tracks/ Lanes Source
Tunis‐ LRT LRT 13,400 Underground Junctions 1 ?
Calgary LRT 5,900 Surface 1 +Portland MAX Blue Line LRT LRT 4,741 Surface 1 *Denver Central Corridor LRT LRT 4 484 Surface 1 *
LRT
Passengers per hour per peak direction observed: BRT, LRT, HRT
Cali BRT 11,100 Surface 2 #Ottawa West Transitway BRT 11,100 Surface 1 &Curitiba‐ Eixo Sul BRT 10,640 Surface 1 &Xiamen BRT 8,360 Elevated 1 #Brisbane BRT 7,700 Surface 2 #Mexico City BRT 7,550 Surface 1 #Zhengzhou BRT 7,230 Surface 1 #Urumqi BRT 6,230 Surface 1 #
Denver Central Corridor LRT LRT 4,484 Surface 1 *Edmonton LRT 3,800 Surface 1 +Phoenix Metro LRT LRT 2,985 Surface 1 *Pittsburgh "The T" LRT LRT 2,017 Surface 1 *Toronto Spadina LRT 2,000 Surface 1 +Newark LRT 1,800 Surface 1 +Sacramento LRT 1,500 Surface 1 +Charlotte Lynx LRT LRT 1,000 Surface 1 *Denver Southwest Corridor LRT LRT 1 268 Surface 1 *Chengdu BRT 6,650 Elevated 1 #
Lanzhou BRT 6,550 Surface 2 #Dalian BRT 6,430 Surface 1 #Hangzhou BRT 6,300 Surface 1 #Quito BRT 6,000 Surface 1.5 #Johannesburg BRT 4,510 Surface 2 #Hefei BRT 3,600 Surface 1 #Yinchuan BRT 3,600 Surface 1 #
Denver Southwest Corridor LRT LRT 1,268 Surface 1 *Seattle South Lake Union (SLU) Streetcar LRT 214 Surface 1 *Portland Streetcar LRT 814 Surface 1 *
Hong Kong‐ Subway HRT 84,000 Underground 2 ?São Paulo‐ Line 1 HRT 60,000 Underground 2 ?NYC Green Lines Combined HRT 56,100 Underground 2 +Santiago‐ La Moneda HRT 36,000 Underground 1 ?
HRT
Jakarta BRT 3,400 Surface 1 #Beijing BRT 2,750 Surface 1 #Changzhou BRT 2,650 Surface 1 #Los Angeles Orange Line BRT 2,357 Surface 2 *Jinan BRT 2,050 Surface 1 #Leon, MX BRT 1,950 Surface 1 #Pittsburgh Martin Luther King, Jr. East Busway BRT 1,714 Surface 2 * Lianyungang BRT 1,650 Surface 1 #
NYC 4,5, express trains HRT 30,200 Underground 1 +Toronto Spadina HRT 26,200 Underground 1 +Manila‐ MRT‐3 HRT 26,000 Elevated 1 ?NYC 6 train HRT 25,900 Underground 1 +London‐ Victoria Line HRT 25,000 Underground 1 ?Montreal HRT 24,400 Underground 1 +Bangkok‐ SkyTrain HRT 22,000 Elevated 1 ?Buenos Aires‐ Line D HRT 20,000 Underground 1 ?
Lianyungang BRT 1,650 Surface 1 #Zaozhuang BRT 1,400 Surface 1 #Yancheng BRT 1,300 Surface 1 #Ahmedabad BRT 1,200 Surface 1 #Bangkok BRT 1,200 Surface 1 #Nantes BRT 1,200 Surface 1 #Las Vegas Strip & Downtown Express (SDX) BRT BRT 1,199 Surface 1 *Cleveland HealthLine BRT BRT 1,129 Surface 1 *Eugene Emerald Express Green Line (EmX) BRT BRT 714 Surface 1 *
Newark Path HRT 17,800 Underground 1 +Washington DC Red HRT 12,700 Underground 1 +Chicago Red HRT 11,900 Elevated 1 +San Fran BART HRT 6,200 Underground 1 +Atlanta HRT 5,100 Underground 1 +Sources:
* Extrapolated from daily demand collected by ITDP for "More Development " https://www itdp org/more developmen
+ Taken from TCQSM 2nd Ed, Annex. http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp100/part%205.pdf, p. 5 ‐ 123
Eugene Emerald Express Green Line (EmX) BRT BRT 714 Surface 1 *Pittsburgh South Busway BRT BRT 662 Surface 1 *Pittsburgh West Busway BRT BRT 601 Surface 1 *Las Vegas Metropolitan Area Express (MAX) BRT 529 Surface 1 *
? BRT Planning Guide, 2007 (cant find original source)
& Taken frm TCQSM 2nd Ed. Part 2: Transit in North America, p. 2‐13# Counted by ITDP China staff, from: http://www.chinabrt.org/en/cities/param‐quan.aspx?param=2
* Extrapolated from daily demand collected by ITDP for More Development… , https://www.itdp.org/more‐developmenfor‐your‐transit‐dollar‐an‐analysis‐of‐21‐north‐american‐transit‐corridors/ using a ratio of 1/14 derived from TCQSM 2nd Ed data
BRT systems with express Corridor Type Speed (km/hr) Source
Pittsburgh West Busway Pennsylvania, BRT 54 [ii]
Comparative Observed Speeds, BRT, LRT, HRT
BRT
services (passing lanes) had higher speeds than
LRT
Pittsburgh Martin Luther King, Jr. East Busway BRT 54 [x]Pittsburgh South Busway, Pennsylvania, BRT 54 [iii]Ottawa Transitway, Canada BRT 52 [i]Orange Line, Los Angeles BRT 32 [v]Bogotá, Colombia, TransMilenio BRT 27 [x]Curitiba, Brazil, Linha Verde BRT 25 [x]Beijing (Lines 1, 2, 3, 4) BRT 24 [iv]Ahmedabad, India, Janmarg BRT 24 [x]
Otherwise speed explained by stop distances and BRT
Ahmedabad, India, Janmarg BRT 24 [x]Guangzhou, China, GBRT BRT 23 [x]Las Vegas Metropolitan Area Express (MAX) BRT 22 [x]Curitiba, Brazil, RIT corridors BRT 18 [x]Los Angeles OrangeLine BRT 18 [x]Cleveland HealthLine BRT 18 [x]Mexico City, Mexico, Insurgentes BRT 17 [x]Eugene Emerald Express Green Line (EmX) BRT 17 [x]
LRTdistances, and BRT Standard elements
(dedicated ROW, at level boarding, off board fare
Sound Transit Central Link, Seattle, Washington, USA LRT 40 [vi]
Ottawa O-Train LRT 40 [x]LYNX Blue Line, Charlotte, North Carolina, USA LRT 37 [vi]Portland MAX Blue Line LRT LRT 30 [x]Denver Central Corridor LRT LRT 23 [x]Denver Southwest Corridor LRT LRT 23 [x]Phoenix Metro LRT LRT 19 [x]
LRT
boarding, off board fare collection, etc.)
Phoenix Metro LRT LRT 19 [x]Budapest, Hungary, Grand Boulevard LRT LRT 18 [x]Portland Streetcar LRT 16 [x]Seattle South Lake Union (SLU) Streetcar LRT 8 [x]
Manila MRT 3 (Metrostar Express), Philippines HRT 48 [viii]
Expo/Millennium Lines, Vancouver, Canada HRT 43.5 [vii]
Tren Urbano San Juan Puerto Rico HRT 33 2 [ix]
HRT
Tren Urbano, San Juan, Puerto Rico HRT 33.2 [ix]
Sources:
[ii] US Department of Transportation. Evaluation of Port Authority of A llegheny County's West Busway Bus Rapid Transit Project . Washington DC,
2003. Report No. FTA-PA-26-7010-03.1 http://www.fta.dot.gov/documents/Pittsburgh_West_Busway_BRT_Evaluation-April_2003.pdf[iii] "Pittsburgh, Pensylvania South, East, and West Busways." Transportation Research Board.
http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp90v1_cs/Pittsburgh.pdf
[iv] National BRT Institute. "Perspectives on Bus Rapid Transit (BRT) Developments in China." Presentation. National BRT Institute. May 1, 2006. http://www.nbrti.org/docs/pdf/Darido_China BRT_051106_presentation.pdf.Transportation Research Metro Orange Line BRT Project Evaluation Vol 0004 Washington DC: Federal Transit Administration 2011
[i]Ottawa Ontario BRT Case Study." Transportation Research Board. http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp90v1_cs/Ottawa.pdf.
Transportation Research. Metro Orange Line BRT Project Evaluation . Vol. 0004. Washington DC: Federal Transit Administration, 2011.
http://www.fta.dot.gov/documents/FTA_Research_Report_0004_FINAL_2.pdf
[vi] Speeds for Charlotte Lynx, Central Link, and WMATA Silver line calculated from posted schedules.
[vii] "Vancouver SkyTrain—A Proven Success Story." Japan Railway & Transport Review, no. 16 (1998): 44‐45. http://www.jrtr.net/jrtr16/pdf/f44_vancouver.pdf.[viii] Antiporda, Jefferson. "DOTC Promises Better MRT Service next Year." Manila Times, August 15, 2014. http://www.manilatimes.net/dotc‐promises‐better‐mrt‐service‐next‐year/119302/.
[ix] "Subways of Puerto Rico." My Transit Guide. http://mapa‐metro.com/en/Puerto Rico/San Juan/San Juan‐Tren‐Urbano‐map.htm.[x] "More Development for your transit dollar", 2013 ITDP. https://www.itdp.org/more‐development‐for‐your‐transit‐dollar‐an‐analysis‐of‐21‐north‐american‐transit‐corridors/, derived from interviews with transit authority staff.
Big emphasis on service planningBig emphasis on service planning
• Service plans determine new system’sService plans determine new system s demand, the fleet size needed, the existing routes affected the speed and capacity of theroutes affected, the speed and capacity of the system, the needed sizing of the system
• Yet most BRT design is done without• Yet…most BRT design is done without reference to a service plan, and service plan has to accommodate whatever infrastructurehas to accommodate whatever infrastructure was designed .
Intermediate
Terminals
transfer stations
Trunk‐feeder services
Direct services
When there was less experience on BRT…
Mexico City
Good but expensive connection tunnels ($40
million+) 1km
TransJakarta did not consider routes going from Blok M (south) to Pulogadung (East), or other options. As a result, a huge transfer bottleneck occurred at
HarmonyHarmony
Kota StationKota StationKota StationKota StationKota StationKota StationKota StationKota StationKota Station
HarmoniHarmoniHarmoniHarmoniHarmoniHarmoniHarmoniHarmoniHarmoniMonasMonasMonasMonasMonasMonasMonasMonasMonas
Kalideras
Pulogadung
Bunderan SenayanBunderan SenayanBunderan SenayanBunderan SenayanBunderan SenayanBunderan SenayanBunderan SenayanBunderan SenayanBunderan Senayan
Blok MBlok MBlok MBlok MBlok MBlok MBlok MBlok MBlok M
Harmoni Station, Jakarta
But over time, methods evolve
Harmoni bottleneck partially resolved by adding second sub‐stop d i l b t di till j bland passing lane but crowding still a major problem.
Rea Vaya BRT, Johannesburgg
A hybrid of trunk/feeder and direct (complementary) services
560 Route using corridor /route length more than 30％296
210
Route using corridor /route length more than 30％
561 242
Flexible Operation Turning buses just leave busway and enter general traffic. Openings in the physical barrier
BRT routes at Shidajida ‐ Guangzhou
New Service planning questions danswered
• How many and which existing routes toHow many and which existing routes to incorporate into a BRT system service plan?
• When to split a direct service into a trunk• When to split a direct service into a trunk route and a feeder route?Li i d i• Limited stop services– When should stops be eliminated, – When should limited stop services be introduced – With what stopping pattern?
What existing routes to include in a lnew BRT Service Plan?
• All of them, except…• Services with so little overlap with the BRT trunk infrastructure the cost of a new bus cannot beinfrastructure the cost of a new bus cannot be recouped with the time savings benefit
• Services outside the administrative authority of ythe regulator of the BRT infrastructure– Intercity buses, charter buses, informal minibuses with bus types incompatible with BRT trunkwith bus types incompatible with BRT trunk infrastructure
• Services most likely to saturate the BRT trunk idcorridor
Rank routes based on “maximum passengers on a route that passes station ‘I’ for the
least amount of total dwell time that route uses at the bottleneck station ‘i’.
Keep adding routes until the addition of the last route slows the total passengers inside the busway down more than the value of
the benefit to the new passengers using the busway.
Priority = Pax(i)/Td(i) y ( ) ( )
Where:Where: Pax (i) is the Load on the buses of a specific bus route passing
the bottleneck station ‘I’. To(i) is the total dwell time in seconds per bus on that route.
Advantages of Direct S i
Advantages from Trunk d dServices
• Less fleet needed as a • Optimizes bus size and
and Feeder Services
Less fleet needed as a result of peak effect.
• Avoids very significant
Optimizes bus size and type
• Can reduce station y gtransfer delays and costs
platform saturation• May improve regularity y p g yof service on trunk
Trunk and Feeder system requires more total fleet than Direct Service Option
Fleet size is set by the maximum peak load for a specific cycle time.
In normal operating conditions, the fleet needed to serve two 1‐hour cycle time routes (one trunk and one feeder) is more than the fleet needed to serve one
Original Formula:
routes (one trunk and one feeder) is more than the fleet needed to serve one two hour cycle time:
Fleet = (Lmhour * TC) / Cbhour b
Revised formula:Revised formula:
Fleet = Loadmax cycle time/Cb
Load max(TC=1) = 63+69+67+66 = 265Loadmax(TC=2) =Loadmax(TC 2)
51+63+69+67+66+53+45+34 = 448
15 Minute Accumulated Peak Period Critical Link Demand Profile
Time Loads load Lm(TC=1) Lm(TC=2)6:00 15 15 118 3836:15 21 36 166 4216:30 31 67 214 4456:30 31 67 214 4456:45 51 118 250 4487:00 63 181 265 4297:15 69 250 255 3877:30 67 317 231 3397:30 67 317 231 3397:45 66 383 198 2938:00 53 436 164 2488:15 45 481 132 2168:30 34 515 108 1928:30 34 515 108 1928:45 32 547 95 1799:00 21 568 84 168
Lm(Trunk) (TC=1) = 265Fleet = 265/60 = 4.417 Lm(TC=2) = 448
Fleet: Direct Service Option Fleet: Trunk and Feeder Option
/Lm(Feeder) (TC=1) = 265Fleet = 265/60 = 4.417
Total Fleet: 8.84
600
( )Fleet = 448/60 = 7.468
500
600
400
15 Minute Loads
Lm(TC=2) = 448Fleet = 448/60 = 7.468
200
300 Accumulated load
Lm(TC=1)
Lm(TC=2)
Lm(TC=1) = 265
100
Lm(TC 1) 265Fleet = 265/60 = 4.417
06:00 6:15 6:30 6:45 7:00 7:15 7:30 7:45 8:00 8:15 8:30 8:45 9:00
Original Services on BRT Corridor
Trunk and Feeder Alternative for BRT C idCorridor
Potential Benefit from Bus size optimization
Higher Potential Benefit from Trunk – Feeder Service
Bus size optimization
g
Less Potential Benefit from Trunk – Feeder Service
Two conditions where Trunk Feeder benefits: 1 A high % of the total route operates along the trunk route so most of the passengers1. A high % of the total route operates along the trunk route so most of the passengers benefit from the efficiencies of larger bus size2. The total demand in the corridor is divided up among a large number of routes, each with relatively low frequency so long waiting times. In these conditions, Trunk and Feeder functions like small businesses joining a cooperative to reach returns to scale. Trunk feeder becomes like a ‘black hole’, the more that join, the more the benefits.
Include which routes in Trunk Feeder operation to optimize bus size? Sort by fewest “Places” brought by each route to the Trunk
(Pl = Lm(TC)max= Lm(hour) * TC)( ( ) ( ) )
The low demand routes have more to gain by joining a ‘co‐op’ for the trunk portion of the route (they can use bigger buses for more of the route).
As more routes join the trunk, bus size on trunk increases as does frequency.
Scenario feederoriginal route Lm Pl Trunk
bus size frequency
Trunk Cost Feeders
Direct Routes total
trunk accumulated costscosts (COF+Cw)
trunk 0 0 0 11346 11346C 529 680 100 130 18.2 5 427 529 10666 11622B 443 749 300 390 31.6 9 739 972 9917 11629F 335 917 700 910 48 3 15 1129 1307 9000 11437F 335 917 700 910 48.3 15 1129 1307 9000 11437A 529 1005 1100 1430 60.5 18 1416 1837 7995 11248D 648 1154 1600 2080 73.0 22 1708 2485 6841 11034E 670 1381 2400 3120 89.4 27 2091 3155 5461 10707G 1059 1606 3200 4160 103.2 31 2415 4214 3855 10484H 3349 3855 5200 6760 131.6 40 3078 7563 0 10641
total 7563 11346
With very high Trunk Route cycle times as a share of the total route bus costsWith very high Trunk Route cycle times as a share of the total route, bus costs continue to fall as more routes are converted to Trunk and Feeder.
With a very short Trunk Route cycle time, bus costs continue to rise as more routes are converted to Trunk and Feeder
With l ti l h t t k t f ll ft t i iti l f tWith a relatively short trunk, costs fall after a certain critical mass of routes have been converted to Trunk and Feeder, but the cost reductions never drop below the cost of the original direct service.
With long Trunk Route cycle times, and high demand on the two routes continuing far beyond the end of the corridor, costs fall as all but the two highest demand and longest routes are added
8200
8300
d)
trunk feeder system‐ coverage and total cost
7800
7900
8000
8100
ng +vehicle fixe
7400
7500
7600
7700
costs (waiti
none B C F A D E G H
sucessive routes entering the trunk feeder system
Costs of the new transfer: Indirectness fof route
• Delay #1:Walking and waiting related
to the new transfer.Internal terminal design should minimize this delayshould minimize this delay.
• Delay #2The transfer terminal is
unlikely to be in a locationunlikely to be in a location optimal to all routes. Some routes will be forced off their original route.
Delay #3. T i l l lik l t b l t d di tlTerminal also unlikely to be located directly on the trunk corridor due to lack of land
availability.
ExamplesExamplesterminal
500 m
s
500 m
terminal
feed
ers
Empirical data on these delays. Benefits of Bus size optimization will rarely be greater than these additional p y g
costs + additional fleet requirements
T i l R f Additi l T i l D l
Description minimum average maximum
Typical Ranges of Additional Terminal Delaysbus & passenger delay
iRerouting to Terminals 0 4 10Bad Terminal Location 0 6 15l d l l
minutes
Internal and External Circulation 2 4 6Additional Boarding & Alighting Delay 1 3 5
Passenger Delay OnlyWalking inside terminal 0 2 6
Additional waiting due to transfer 2 4 6Total time/ passenger: minutes 5 23 48
Equivalent time/passenger: minutes 7 31 66
Station platform saturation may be the f k d dmajor cause of Trunk and Feeder use
• Direct services leave passengers waiting on the station platform longer (frequency per route is lower
lso passengers accumulate on the platform.
• If the platform width is il blunavailable, some routes may
be converted to trunk and feeder routes to relocate the necessary platform space to anecessary platform space to a transfer terminal where land is more readily available.
Stop removal
Optimal distance between station: l b id 450normal urban corridor 450 meters
30
35
15
20
25
me
(min
utes
)
0
5
10Tim
In-vehicle timeWalking time
Total travel time
200 300 400 500 600 700 800 900 1000 1100 1200
Distance (metres)
Remove stops with low
code location boarding alighting code location boarding alighting14182 Western & Berwyn Terminal 3 0 0 6648 Western & Ogden 49 40 214529 Western & Berwyn 1013 1 0 8245 Western & 14th Street 11 21 11717 W t & F t 390 14 0 14531 W t & 16th St t 41 32 0
bus stop daily demand elimination code
bus stop daily demand elimination code
with low demand, or if
1717 Western & Foster 390 14 0 14531 Western & 16th Street 41 32 01718 Western & Carmen 51 0 1 8249 Western & 18th Street 27 40 11719 Western & Winnemac 66 4 0 8250 Western & 19th Street 28 38 21720 Western & Ainslie 73 12 2 15059 Western Pink Line Station 278 221 014591 Western & Lawrence 513 73 0 14555 Western & Cermak 350 393 08184 Western & Leland (Brown Line) 419 206 0 17058 Western & 23rd street 38 57 18185 Western & Wilson 83 16 2 8255 Western & 24th Street 88 56 08186 Western & Sunnyside 36 17 1 8256 Western & 25th Street 19 14 18187 Western & Montrose 326 133 0 8257 Western & 26th Street 190 184 08188 Western & Cullom 29 22 2 8258 Western & 27th Street 24 24 18190 Western & Belle Plaine 20 46 1 8259 Western & 28th Street 60 30 214964 Western & Irving Park 341 133 0 8260 Western & 31st Street 17 11 0
stops simply too close together
8192 Western & Byron 14 5 1 8262 Western & 33rd Street 4 27 18193 Western & Grace 70 49 2 8263 Western & 34th Street 10 26 08195 Western & Addison 512 358 0 8264 Western & 35th Street 78 131 08196 Western & Cornelia 68 128 1 8265 Western & 36th Street 11 31 18197 Western & Roscoe 157 101 0 16085 Western & Archer 171 268 08198 Western & School 91 33 1 14575 Western & Pershing 36 36 18199 Western & Belmont 391 233 0 8268 Western & 40th Street 18 29 08200 Western & Barry 37 10 1 8269 4102 S Western 16 14 18202 Western & George 31 51 2 8270 Western & 42nd Street 32 54 014585 Western & Elston/Diversey 323 260 0 8271 Western & 43rd Street 92 95 18204 Western & Schubert 99 111 1 8272 Western & 44th Street 21 53 08205 Western & Logan/Jones 283 68 0 15743 Western & 45th Street 33 52 1close together,
if there is h
8206 Western & Altgeld 352 259 0 8274 Western & 46th Street 27 61 28207 Western & Fullerton 86 42 2 8275 Western & 47th Street 315 445 08208 Western & Belden 68 35 1 8276 Western & 48th Street 12 13 18210 Western & Palmer 49 54 2 14132 Western Orange Line Station 860 439 08211 Western & Charleston 126 215 0 8278 Western & 50th Street 4 19 28212 Western & Armitage 184 244 1 8279 Western & 51st Street 92 155 08213 Western & Milwaukee 431 271 0 8280 Western & 52nd Street 5 34 18214 Western & Cortland (Blue Line) 389 219 0 8281 Western & 53rd Street 75 99 08215 Western & St. Paul 26 19 1 8282 Western & 54th Street 55 77 18216 Western & Wabansia 57 98 2 8283 Western & 55th Street/Garfield 274 282 014581 Western & North Avenue 344 288 0 8284 Western & 56th Street 65 66 117343 Western & Le Moyne 70 126 2 15143 Western & 57th Street 56 70 0
another station in a reasonable
y8219 Western & Hirsch 87 118 0 8286 Western & 58th Street 27 89 18220 Western & Potomac 42 93 1 14578 Western & 59th Street 135 165 08221 Western & Division 407 598 0 8288 Western & 60th Street 24 54 18222 Western & Thomas 46 106 1 8289 Western & 61st Street 76 106 08223 Western & Augusta 90 212 0 8290 Western & 62nd Street 62 179 08224 Western & Iowa 34 109 1 14576 Western & 63rd Street 476 441 08225 Western & Chicago 374 491 0 8292 Western & 64th Street 38 65 18226 Western & Huron 14 17 1 14857 Western & 65th Street 80 124 08227 Western & Ohio 23 36 0 8294 Western & 66th Street 35 116 18228 Western & Grand 125 104 1 14823 Western & Marquette Rd 99 200 015344 Western & Hubbard 95 66 2 8296 Western & 68th Street 15 93 18230 Western & Fulton 31 81 1 10117 Western & 69th Street 147 394 0
distance8230 Western & Fulton 31 81 1 10117 Western & 69th Street 147 394 08231 Western & Lake 56 91 0 8298 Western & 70th Street 17 61 18233 Western & Warren 95 131 1 8299 Western & 71st Street 154 336 014546 Western & Madison 364 241 0 8300 Western & 72nd Street 35 92 18236 Western & Adams 96 68 1 15783 Western & 73rd Street 27 98 08237 Western & Jackson 92 147 0 8302 Western & Columbus 48 74 28239 Western Blue Line Station 499 282 0 8303 7521 S Western 5 12 114769 Western & Harrison 110 116 0 8304 Western & 76th Street 5 13 017312 Western & Polk 59 111 2 8305 Western & 77th Street 1 17 115345 Western & Taylor 79 143 0 8306 Western & 78th Street 0 626 0
When to add express routesWhen to add express routes
• When the removal of fixed dwell time (perWhen the removal of fixed dwell time (per stop) benefits more riders than the additional delay caused by lower frequency of servicedelay caused by lower frequency of service
• When adding a service reduces the irregularity of boarding and alighting delayof boarding and alighting delay. – Boarding and alighting delay becomes irregular and causes bunching at frequencies of greaterand causes bunching at frequencies of greater than 30 per hour, with 22 per hour optimal.
Two conditions where adding limited bstop or express services is obvious
• When demand is so high on a single route thatWhen demand is so high on a single route that the frequency is greater than 30 buses/hour. In this case the addition of a limited stopIn this case, the addition of a limited stop service will yield benefits in almost all cases.
• When demand is highly concentrated in a• When demand is highly concentrated in a limited number of station stops.
If demand is relatively uniform, the below i i ll i lstopping pattern is usually optimal.
8000
9000
express route ‐ example of demand boarding
alighting
3000
4000
5000
6000
7000
8000
pass/hou
r
Load
0
1000
2000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
p
normal express normal
A B CA B CPassengers traveling within zone A or within Zone C can take local or express so no loss of benefit. So for 2 OD pairs there is no change.
Passengers traveling from A to C or C to A can all take Express and gain Td * 7 stops. So for 2 OD pairs, there is a big time savings benefit.
Passengers traveling between A and B or B and C lose 50% of frequency orPassengers traveling between A and B or B and C lose 50% of frequency, or waiting time * 2. So for 2 OD pairs there will be a disbenefit.
The outcome depends on the demand profile
Empirical evidence shows Benefits of express routes tend to be higher for Trunk and Feeder systems (on average in the 40%higher for Trunk and Feeder systems (on average in the 40% range) than for Direct Service systems (in the 20% range) because demand is concentrated at the terminal and the
downtown
12000
express route ‐ example of demand
6000
8000
10000
ss/hou
r
boardingalighting
0
2000
4000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
pa
normal express normal
Load
A B C
Optimal pattern likely to be two services: one stopping everywhere and one stopping only at the two extremes, the terminal and downtown
If demand is very high, more patterns y g , pcan probably be sustained keeping the frequency per route in the optimal 15frequency per route in the optimal 15
– 30 range. This is the reason TransMilenio services have this look
Express 3
E 2Express 2
Express 1
Local
Early return services allow for higherEarly return services allow for higher frequency on high demand portion with the same fleet or reduction in fleetthe same fleet, or reduction in fleet
and
nand
nen
ger d
emr d
ireci
tion
B
A
enge
r dem
r dire
citio
n
B
A
Pass
epe
B
Pass
epe
B
Travel time along corridorRT
Travel time along corridorRT
l l b fSplitting a longer route is a combination of two early returns.
and
nand
nen
ger d
ema
r dire
citio
n
B
A
enge
r dem
ar d
ireci
tion
B
A
Pass
epe
r B
Pass
epe
r B
Travel time along corridorRT
Travel time along corridorRT
An example from YichangAn example from Yichang
Actual itinerary for routes 101,102 Proposed itinerary for routes 101 102 & 103 cut integrating iny ,
& 103101,102 & 103 cut integrating in
BRT corridor
Decreasing load from a transfer t i l d t E l fterminal or a downtown: Example of Santa Amaro Corridor in Sao Paulo
14000
16000
example of continuous decrescing load boarding
8000
10000
12000
pass/h
alighting
load
0
2000
4000
6000
01 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
stations
Optimal pattern turns out something like this: (also likely to be used onlike this: (also likely to be used on
TransBrasil BRT)
12000
13000
14000
15000
16000demand and express routes
boarding
alighting
load
Local.1
exp 1
7000
8000
9000
10000
11000
pass/h
exp 1
exp 2
exp 3
local
1000
2000
3000
4000
5000
6000
0
1000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
stations
This combines clustered stop removal with early return services. Fleet does not need to go all the way to the end, so there is a big reduction in the needed fleet.
New materials on institutional structures for BRTstructures for BRT
Transportation Authority (transit and traffic)
Transit Authority (rail also)
BRT and Other Buses and Minibuses
Special Purpose BRT Agency
Public Bus operator
Transport Department
BRT Management Authority Options
TransMilenio, Bogota X (Phase IV) X (Phases I ‐ III)GBRT, Guangzhou XBRT, Curitiba XGuadalajara, Mexico XDar es Salaam XLima BRT XDar es Salaam (pending) X*a es Sa aa (pe d g)Perreira, Colombia XMio, Cali, Colombia XJan Marg, Ahmedabad XRea Vaya, Johannesburg XMetrobus, Mexico City XHealthLine, Cleveland XLanzhou BRT XLanzhou, BRT XTransJakarta, Jakarta XSao Paulo, Brazil XLAMTA (Orange Line) XCape Town BRT XPittsburgh ACTA XLondon (No BRT in London) X
( )San Francisco (no BRT yet) XDakar (CETUD) XBRT basic, Lagos X
Gold Standard BRTSilver Standard BRTBronze Standard BRTBasic BRT or Below
MunicipalityMunicipality(Mayor)
Metro Company Transportation BRT Authority Traffic PolicePublic Works
Metro Company Department
Transit licenses
BRT Authority
BRT Operations
Traffic Police
Station security
(BRT Infrastructure)
Traffic management Bus operating Traffic signals Station Architects,
Road Engineering
Fare collection ConstructionAdvantages and disadvantages of
Operational Control
alternative structures explained
Advantaes and disadvantages of l dcontracting out options explained
Curitiba Bogota Santiago Jakarta Johannes Cape Ahmed Guang Mexico ITDPCuritiba Bogota Santiago Jakarta Johannes-burg
Cape Town
Ahmedabad
Guang-zhou
Mexico City
ITDP Recommended
Urbs Trans-Milenio
Trans- Santiago
Trans-Jakarta
Rea Vaya MiCity Jan Marg
GBRT Metro-bus
(Africa)Milenio Santiago Jakarta Marg bus
Bus Procurement Private Private Private Public Private Public Private Private Private PrivateBus Operations Private Private Private Private Private Private Private Private Private PrivateF C ll ti P bli P i t P i t P i t P i t P i t P bli P bli P i t P i tFare Collection Public Private Private Private Private Private Public Public Private PrivateTrust Fund Public Private Private Public Public Public Public Public Private PrivateControl Center Public Public Private Public Public Public Public Public Public PrivateO ti l Pl i P bli P bli P i t P bli P bli P bli P bli P bli P bli P i tOperational Planning Public Public Private Public Public Public Public Public Public Private
Examples of BRT Systems with Multiple Private BRT BusMultiple Private BRT Bus
Operators
BRT System Phase I Operators Phase II operatorsNumber of Operating Companies
Bogota 6 8Guangzhou 3 3Curitiba 2 4Curitiba 2 4Mexico City 2 3Rea Vaya 1 2Ah d b d 1 1Ahmedabad 1 1MiCity 1 3TransJakarta 1 2
BRT Standard Rank Not BRTGoldSilverBronze
Corporatization is key to Gold BRT
Former Mi ib
Mixed F
Former Private Bus C i C ll i
BRT Operator Ownership Type
Minibus Operators Formed into Companies
Former Minibus and Private Investors
Companies under new contract form
Outside Private Investor
Collective without Integrated fleet Management
Public bus operator
TransMilenio Bogota X XTransMilenio, Bogota X XGBRT, Guangzhou XBRT, Curitiba XGuadalajara, Mexico X XLima BRT X XPerreira, Colombia X XMio, Cali, Colombia X XJan Marg, Ahmedabad XRea Vaya, Johannesburg XM b M i Ci X XMetrobus, Mexico City X XTransJakarta, Jakarta X XHealthLine, Cleveland XLanzhou, BRT XLAMTA (Orange Line) XLAMTA (Orange Line) XMiCity, Cape Town X XLiteBRT, Lagos XPorto Alegre Basic BRT X
provide feeder services in any BRT system. Some feeder services usesystem. Some feeder services use
minibuses. Important not to confuse th hi l t f th t t tF d B O tithe vehicle type from the contract type.
Former minibus operators formed into
Same companies No Feeder, No Feeder,
Feeder Bus Operations
formed into formal companies
companies as Trunk Operator
No Feeder, Direct Services
No Feeder, Trunk Services Only
Informal minibus feeders
TransMilenio, Bogota XGBRT Guangzhou XGBRT, Guangzhou XBRT, Curitiba XJan Marg, Ahmedabad XMio, Cali, Colombia XR V J h b XRea Vaya, Johannesburg XMetrobus, Mexico City XTransJakarta, Jakarta XHealthLine, Cleveland XLanzhou, BRT XLAMTA (Orange Line) XMiCity, Cape Town XPorto Alegre Basic BRT X