Click for next page ( 6


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 5
have been used in this chapter to examine travel behavior responses and related impacts of HOV facilities. Only a few HOV facilities have been the subject of ongoing comprehensive assessments. In most cases, assemblage of data over time introduces questions about consistency, and informa- tion on many potential travel behavior influences may be spotty. Judicious utilization of data pre- sented is obviously called for. A limitation deserving special note is the scarcity of available analyses that examine HOV facility effects over the full lateral extent of travel corridors. Much available research is operationally ori- ented and focused on HOV lanes themselves. In some cases even the travel demand information on the overall highway facility within which an HOV lane operates may be incomplete. These con- straints often make definitive generalizations about broad impacts difficult. It is frequently neces- sary to simply assume that if a facility operates more efficiently with an HOV lane, the corridor probably does also. The corresponding implications as well as related issues affecting assessment of air quality and environmental impacts of HOV lanes are discussed under "Impacts on Energy, Air Quality, and Environmental Factors" in the section on "Related Information and Impacts." Comprehensive data that does exist is often old, some of it dating to the late 1960s through 1970s period of initial bus and HOV lane experimentation. Of course, individual HOV facilities open only once, so data that describe the initial traveler response to an operation in place for some time will necessarily be "old data." Much basic HOV lane inventory data is circa 1998. This circumstance is not as much of a problem as it might seem, because usage of many facilities--particularly the bet- ter established ones--has been relatively stable throughout the late 1990s and early 2000s. This can be seen in several late 1990s versus 2004 data comparisons. The different definitions possible for Average Vehicle Occupancy (AVO) introduce potential for inconsistency. AVO statistics may be calculated exclusive or inclusive of bus vehicles and their passenger loads, a difference that becomes enormously important on facilities with any signifi- cant bus use. Where information allows, AVO statistics are specifically identified as to whether they are auto (including carpool) and vanpool; carpool and vanpool; auto, vanpool and bus; or carpool, vanpool and bus AVO. When the type of AVO cannot be explicitly identified, or where the term "auto occupancy" is used, it is most likely to be auto and vanpool AVO (i.e., without buses included). On the other hand, unidentified AVOs a unit or more higher than the minimum carpool occupancy requirement can with some assurance be assumed to include buses. Clearly the reports of AVO need to be used with extra caution in instances where the type of AVO can- not be assured. Still another consistency problem involves the issue of whether violators should be counted in HOV lane vehicular and passenger volume statistics or not. In most cases, it is simply not known which was done by reporting agencies and authors. In the few instances where violators were sep- arately identified, they have mostly been included in the HOV lane volume counts. Exceptions are noted. TRAVELER RESPONSE SUMMARY The attractiveness of HOV facilities and traveler response to them depends on the travel time they save for the user, the trip time reliability afforded, the types and levels of bus service on the facil- ity, location and orientation within the urban area, HOV lane use eligibility requirements, years in 2-5

OCR for page 5
service, presence of supporting elements such as park-and-ride lots, and corridor congestion lev- els. Aside from the fundamental differentiation between freeway and arterial HOV facilities, type of facility per se is not a major determinant of attractiveness, nor is facility length, except as an indi- cator of how much congestion is bypassed. The U.S. HOV facilities with the highest passenger vol- umes include a reversible exclusive facility, a set of concurrent flow lanes, and a contraflow lane. Size and configuration of urban area population and major employment centers are critical deter- minants of use. The presence of congestion on the GP lanes and parallel highway facilities is nearly always an essential ingredient of HOV lane effectiveness. Most HOV facilities carry more people per lane than do the adjacent GP freeway lanes in the peak hour, if not the entire peak period. Summarizations of operating results are provided, within the main body of this chapter, in Tables 2-22 and 2-23. Illustrative examples of AM peak-hour vehicle and person volumes on HOV projects include some 500 to 600 buses carrying 23,000 passengers on the NJ Route 495 bus-only contraflow lane approaching the Lincoln Tunnel to New York City; 1,200 vehicles including 22 buses, carrying 3,600 people including 1,100 bus passengers, on the exclusive Northwest HOV lane in Houston; 1,200 vehicles including 64 buses, and 5,600 people including 2,600 bus passengers, on the I-5 North concurrent flow HOV lanes in Seattle; and 1,300 carpools and vanpools with 3,000 occupants on the California Route 91 concurrent flow HOV lanes in Los Angeles County, California. As these figures suggest, HOV lanes may focus on serving buses only, or primarily carpools, or more commonly, a mix of buses, vanpools, and carpools. Projects with the higher bus volumes, which are almost all radial to urban central areas, generally have the higher person movement in the HOV lanes. The average HOV facility carries some 40 percent of its person volume on buses, and total HOV person volumes (bus riders and carpool and vanpool occupants) are closely related mathematically to the number of buses. Central Business Districts (CBDs) are the major source of HOV facility users: 56 percent in the case of the Katy Freeway in Houston, where three major activ- ity centers attract another 22 percent. Two documented examples of lowering freeway HOV lane occupancy requirements suggest that if a lane carries 500 carpools with a 3 occupancy requirement, it may carry on the order of 1,400 with a 2 requirement. The effect on person volumes will depend on bus usage, but it does appear that the greatest person throughput will be achieved with the most liberal lane use eligibility requirements that can be sustained without creating HOV lane congestion. On the other hand, one example based on forecasting and one example based on an actual trial indicate that with peak-hour peak-direction per lane 3 occupancy carpool volumes approaching 1,000 or more, the outcome of lowered occu- pancy requirements is substantial loss of time savings and reliability paired with increased bus costs. Two other examples indicate that user objections to raising the occupancy requirement from 2 to 3 in response to building HOV congestion can be mitigated by allowing 2 carpools on for a toll. Many arterial bus lanes in North America are limited in extent and are more important to bus and traffic operations than as major inducements to transit use, although at least one notable installation--in Manhattan--has resulted in substantial ridership increases at the individual route level. Arterial HOV lanes open to carpools are few in number. One of the more intensively used examples, in Vancouver, BC, carries some 40 buses and 600 to 700 2 carpools in the peak hour, a 38 percent carpooling increase in the initial 7 months. The travel time savings and reliability improvements offered underlie the attractiveness of HOV facilities for users. These benefits may accrue from short queue bypass HOV lanes as well as longer 2-6

OCR for page 5
facilities, particularly at bottlenecks along high-type facilities, or at geographic barriers like water crossings. Documented AM peak-hour travel time savings provided by freeway HOV facilities over traveling on the GP lanes range from a low of 1 minute on the 4-mile CA 57 Orange Freeway HOV lane in Los Angeles County to a high of 37 minutes on the 27-mile I-95/I-395 HOV facility in Northern Virginia/Washington, DC. Time savings on arterial street bus-only lanes depend on the amount of street congestion before lane installation. Improvements in travel time reliability for HOVs appear to be a universal benefit and for all types of facilities include, on average, a halving of "late" bus arrivals. Travel demand model research and surveys both suggest that transit use and conventional carpool- ing do not closely compete with each other in the context of a new HOV facility; both modes draw substantially from single occupant auto use. Results from surveys conducted of bus riders, carpool- ers and vanpoolers on HOV facilities in various U.S. cities indicate that roughly 25 to over 50 percent formerly drove alone, with carpoolers and vanpoolers more toward the upper end of that range. Shifts in carpool, vanpool and bus route choice, sometimes on the order of 15 to 35 percent of HOV lane carpool users, also take place. An exception to normal patterns appears to apply to causal car- pooling, where on established HOV facilities 75 to 95 percent of riders picked up during spontaneous carpool formation report transit or other ridesharing arrangement as their previous mode. Unless pre-existing corridor ridership is high, initial bus passenger volumes on new HOV facili- ties may be a quarter or less of ridership after 2 to 4 years, with the rate of growth depending heav- ily on the program of bus service development. Assuming no changes in occupancy requirements, carpool volumes on new HOV facilities may be about half the volume achieved after a year or two. These are very rough guidelines, as there is wide variability. It is fairly common for HOV lanes ultimately to reach a level of maturity where little or no growth in use is experienced, particularly where introduction of parallel transit or highway improvements compensates for area growth. Growth on some existing facilities slowed markedly after 3 or so years, other facilities sustained steep growth for 6 to 8 years, and at least one plateaued immediately. Although traffic volumes and vehicle miles of travel (VMT) may dip slightly when an HOV lane is opened, and be kept lower than they might otherwise be, HOV facilities do not appear able to counter long-term growth trends in travel demand. A more realistic expectation is that HOV lanes may help reduce growth in VMT and increase potential person carrying capacity by inducing higher vehicle occupancies. The long-term increase in auto, vanpool and bus average vehicle occupancy (AVO) for a freeway where an HOV lane is opened normally occurs within a range of 0.05 vehicle occupants (AVO up 6 percent or so) to 0.25 vehicle occupants (up some 20 percent). I-5 North in Seattle had an AVO increase of 0.45 measured over a 10-year span, from 1.24 to 1.69, a 36 percent increase. A small number of examples have incurred slight occupancy declines. The average AVO increase over a number of facilities is 8 or 9 percent. Corridor-wide effects are more muted where parallel high- ways are in place, perhaps one-half to two-thirds the effect as measured on the freeway. HOV facilities carry about 25 to nearly 40 percent of all freeway person movement in the AM peak hour, peak direction in Los Angeles, Houston, and Minneapolis. These percentages, which are regional averages for freeways with HOV lanes, are above or well above the corresponding aver- age proportions of freeway lanes allocated to HOV facilities. Thus the HOV lane productivity is higher than GP lane productivity overall in these examples. Experience suggests a series of indicators of HOV facility success, most of which should be met for reasonable assurance of satisfactory outcome. Urban area characteristics should desirably include 2-7