U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
This magazine is an archived publication and may contain dated technical, contact, and link information.
|Publication Number: FHWA-HRT-09-003 Date: Mar/Apr 2009|
Publication Number: FHWA-HRT-09-003
Issue No: Vol. 72 No. 5
Date: Mar/Apr 2009
FHWA is exploring innovative ideas for combining congestion pricing and active traffic management to expand highway capacity.
|This actively managed motorway in the United Kingdom (UK) shows the far left lane used as a "hard shoulder" travel lane during peak hours and a refuge area for emergency turnout. Also shown is a gantry with lane controls. (In the UK, motor vehicles drive on the left.)|
In many metropolitan areas across the United States, congestion is clogging freeways, resulting in lengthy travel times and low reliability (that is, people cannot rely on how long it will take to drive from one location to another, making it difficult to plan and predict travel times). To address congestion on highways serving core downtown destinations, transportation agencies have successfully added high-occupancy vehicle (HOV) lanes to support transit and other rideshare services. A few agencies have demonstrated that congestion pricing in the form of high-occupancy toll (HOT) lanes can be an effective way to improve the efficiency of the highway system, while at the same time generating new revenue for transportation investment.
Significant operational, safety, and public acceptance issues, however, constrain the more widespread application of pricing strategies. In particular, while the straightforward conversion of an underutilized HOV lane to an HOT lane is well understood and widely accepted, relatively few such opportunities exist in urban corridors. Deployment of priced lanes in major urbanized areas is likely to be constrained by the limited number of existing HOV facilities, particularly those having excess capacity that could be offered to drivers of single-occupant vehicles willing to pay a price. Other constraints include difficulties in obtaining additional rights-of-way for new lanes, their high construction costs, and public resistance to "take away" of existing toll-free, general-purpose lanes to create priced lanes.
If pricing is to become a more widely used tool to reduce congestion or provide reliable travel times in major metropolitan areas, new approaches to implementation must be developed. One potential solution involves the creation of networks of free-flowing express lanes by (1) using active traffic management (AcTM) strategies to dynamically manage freeways with flexible use of shoulders as travel lanes, and (2) restriping of existing pavement into narrower lanes in order to accommodate a new lane within the existing facility footprint. Agencies would operate the left lane as a priced lane, with the right-side shoulder serving as a general-purpose lane either permanently or when needed to accommodate high demand. This approach avoids the need to take away an existing lane to create the new priced lane. Revenue from the new priced lane could be used to pay for construction of the toll infrastructure, other physical improvements, and the systems needed to actively manage the facility, as well as ongoing operation and maintenance. In some corridors the demand for priced systems might even be sufficient to generate additional revenue for larger capital investments.
These ideas build on emerging strategies now being explored by the transportation community as possible options for providing new highway capacity without the need for new rights-of-way or major reconstruction. The safety and efficiency of these approaches, however, have not yet been fully assessed. In fact, many in the U.S. transportation community have raised serious safety and operational concerns about some AcTM strategies and narrower lane widths. Significant hurdles need to be cleared with respect to public acceptance. Therefore, these new ideas are not ready for immediate application. Rather, the intent is to engender discussion and further exploration through collaboration among the transportation planning, safety, and operations communities to find workable strategies to advance congestion pricing in the United States.
AcTM involves a group of strategies to improve traffic flow and reduce congestion on freeways. Seen as a less expensive alternative to widening a road, AcTM makes use of automatic systems and human intervention to manage traffic flow and ensure the safety of road users. AcTM systems have been implemented successfully in Europe, most notably in the Netherlands and the United Kingdom.
One AcTM strategy involves converting a highway's hard shoulder into a travel lane during periods of high traffic flow in order to expand the road's capacity. A computerized system monitors traffic flows and sets the optimum speed limit for the current traffic flow. Variable speed limit signs may be mounted overhead on gantries, along with lane-control signs over each lane and dynamic message signs (DMS). The DMSs provide advance warning of queues or incidents downstream. Operators monitor closed-circuit television (CCTV) cameras along the route and can control both the speed limits and information signs. The DMSs can direct drivers to use the hard shoulder during busy periods. Only when the speed limit has been lowered to 80 kilometers per hour, km/h (50 miles per hour, mi/h) or below is the hard shoulder permitted to be opened as an additional travel lane.
To facilitate this operation and still maintain safety, a series of emergency turnout areas are created at frequent intervals along the highway. They contain phones for vehicle occupants to call for help. In the event of a vehicle breaking down on the shoulder lane or another travel lane, operators can close the lane using the overhead lane controls. In addition, they can close a lane to allow emergency services access to a crash or other incident. Close to exit and entrance ramps, use of the hard shoulder as a lane is restricted to traffic exiting or entering the highway. Enforcement cameras may be mounted on the gantries to enforce the mandatory variable speed limits.
Although the European experience with AcTM has been positive, agencies in the United States are just beginning to explore this approach. The situation in the United States differs from Europe in significant respects, and some transportation experts in this country have raised safety and operational issues with various aspects of AcTM.
Within the next year, Minneapolis, MN, is expected to make advances in the state of the practice of AcTM in the United States. New highway capacity will be created in Minneapolis on I-35W through use of existing paved shoulders as priced travel lanes during peak hours, along with most of the features of AcTM discussed above, including overhead gantries to provide variable speed limits and lane control for all lanes.
Another key feature of the concept presented here is the restriping of existing lanes that are 3.7 meters (12 feet) wide (the traditional design standard) to create narrower lanes 3.4 meters (11 feet) wide. Narrow lanes are used frequently in work zones. Restriping of existing freeway pavement into narrower 3.4-meter (11-foot) lanes has been implemented on a permanent basis in Los Angeles and Seattle in order to create new HOV lanes without expanding the highway. Since 2007, this strategy has been used on I-90 in Minneapolis, along with partial use of shoulders, to provide extra capacity in the corridor after the collapse of the I-35W bridge. Late in 2008, restriping of lanes was used to create an additional lane on I-95 in Miami, FL. The existing five lanes and shoulders in the northbound direction (including an HOV lane) were restriped to create six lanes, two of which were designated as HOT lanes.
The concept proposed here differs from the Los Angeles, Seattle, and Miami applications in that the narrower 3.4-meter (11-foot) lanes are to be operated in conjunction with AcTM, including overhead DMSs indicating lower (variable) speed limits.
Agencies could restripe freeways as follows:
If extra pavement width is needed, agencies could add pavement or take it from the left shoulder where excess shoulder width exists.
A study by the Midwest Research Institute, Safety Effects of Using Narrow Lanes and Shoulder-Use Lanes to Increase the Capacity of Urban Freeways (FHWA-HRT-05-001), suggests that the use of the added lanes as HOV lanes may introduce a difference in speed between the HOV lane and the adjacent lanes and increase crashes. AcTM on all lanes with speed controls may be helpful in reducing such speed differentials. The I-35W project in Minneapolis, which will have a priced lane adjacent to regular lanes, will use AcTM in part to address this issue. The project will be evaluated extensively. Also, an all-lanes-priced scenario, which results in all vehicles traveling at essentially the same speed (discussed later in this article), may address this issue.
In a 2003 policy statement for the Reason Public Policy Institute, HOT Networks: A New Plan for Congestion Relief and Better Transit, Robert Poole, Jr., director of transportation studies at the Reason Foundation, and Kenneth Orski, editor of Innovation NewsBriefs, presented an HOT concept of a network of express toll lanes using existing HOV lanes and building new HOT lanes. HOVs that meet occupancy requirements would continue to travel toll-free on the HOT lanes, while lower occupancy vehicles would be required to pay a toll.
This plan might not be feasible in some circumstances. "We conducted a rigorous study of the concept and found that such a system would cover less than half the overall capital and operating costs in the Washington [DC] area, due primarily to the high costs for construction of new lanes and interchanges," says Ronald Kirby, director of transportation planning at the Metropolitan Washington Council of Governments. (The report on the 2008 study is Evaluating Alternative Scenarios for a Network of Variably Priced Highway Lanes in the Metropolitan Washington Region.) Also, such a network would take several years to evolve because of the need for extensive new construction.
Proposed here is the expansion of Poole and Orski's idea into a new approach: a combination of converted priced express lanes and AcTM. The express lane network would involve relatively little capital investment and could be implemented in a short time, as it would not require major widening of highways to add lanes.
The proposed concept calls for transportation agencies to restripe metropolitan freeways with full-design-width shoulders to create "dynamic" shoulder lanes adjacent to the existing right lanes. During periods of heavy traffic, drivers could use the shoulder lanes as travel lanes, and agencies would oversee them with AcTM. Agencies could create emergency turnout areas for use when the shoulders are serving as travel lanes. To reduce the time and cost for acquisition of new rights-of-way, the turnouts could be selectively located in areas where rights-of-way are already available.
The agencies then could convert the far left lane to a premium-service express toll lane providing free service only for buses and registered vanpools, and charging a variable toll to all other vehicles -- in effect, an HOT lane with higher occupancy requirements to make vehicle occupancy enforcement easier. To distinguish this variety of toll lane from normal HOT lanes (that permit free use by carpools with as few as two persons per car), and from express toll (ET) lanes (that may not permit free use by vanpools and/or transit vehicles), the premium-service toll lane is termed a flexible and efficient express lane, or FEE lane by the author. The term denotes the flexible enhancement of capacity during rush hours, efficient flow of traffic in the lane, and speedier express travel. The concept essentially would combine three key elements: (1) dynamic use of shoulders as travel lanes to enhance capacity, (2) pricing of left lanes to create a reliably faster trip and control demand, and (3) free service provided only for buses and registered vanpools to simplify vehicle occupancy enforcement.
All vehicles would need electronic transponders to use the FEE lane, and agencies would set toll rates dynamically (in real time) to moderate demand and ensure free flow of traffic while maximizing use of the lane. Dynamic message signs would communicate to drivers the current toll rate and (optionally) the expected travel times on the FEE lane.
Comparison of Vehicle and Person Throughput For Existing Versus FEE Lanes Concept
On a freeway in which the far left lane is already an HOV lane, a transportation agency could add the adjacent lane to create a two-lane FEE section. The number of lanes available for toll-free, general-purpose use would stay the same as before the conversion. During rush hours, the FEE lanes could provide travel time advantages to buses and vanpools certified by employers or the metropolitan ridesharing agency. Carpools could obtain the same advantages at a discount by sharing the cost of the toll. (For example, if there were three persons in a carpool, the toll would be divided between them so that each would pay one-third.) In addition, the FEE lanes could provide reliable trip times for others who absolutely must get somewhere on time and are willing to pay for it.
"With HOT lanes in place, express bus and vanpool services would provide better service to existing customers and would attract more riders due to travel time advantages," says Arthur Guzzetti, vice president of policy at the American Public Transit Association. If new FEE lanes are put in place, express bus services operating in the corridor will be able to provide faster service by using the new free-flowing lanes instead of being stuck in traffic on general-purpose lanes. If FEE lanes are implemented by converting existing HOV lanes, revenue generated from tolls may be used to enhance bus services on the lanes. For example, when HOT lanes were implemented by converting the existing I-394 HOV lanes in Minneapolis, transit ridership increased because park-and-ride facilities were expanded and express bus service frequency was improved as part of the project.
Because all vehicles would have transponders before they can use the FEE lanes, agencies could enforce payment violations by vehicles without valid transponders, using technologies and procedures focused on license plate recognition, already widely employed by toll agencies, along with followup through citations sent through the mail. Enforcing carpool occupancy requirements on normal HOT lanes is difficult, because the shoulder space needed by police vehicles for visual inspection of self-declared carpools may not be available. This challenge is addressed in the FEE lane concept by eliminating the need for onsite police enforcement of occupancy violations. Preauthorized transit vehicles and registered vanpools would obtain special transponders to identify them and provide them with free service. "Since all vehicles would pay, except authorized buses and vanpools assigned special transponder ID numbers, vehicle occupancy enforcement would be greatly simplified," says Poole.
If policymakers want to provide free or discounted service for carpools, enforcement challenges potentially could be addressed by requiring preregistration of eligible carpools (similar to the requirement for vanpools under the normal FEE lane concept). Special transponders would be issued to these preregistered carpool vehicles. The transponders would have a button or switch that could be flipped according to whether the preregistered vehicle is operating as an HOV eligible for free or discounted service. However, this does not eliminate the need to monitor compliance, since a solo driver still could use a preregistered vehicle to avoid paying the toll. To eliminate the need for onsite enforcement, auditing of vehicle occupancies could be done periodically at the employment site, but reliable audit methods would need to be developed.
Although the proposed concept greatly reduces implementation costs, there is a valid concern that it might, at least in this country, introduce significant safety and operational problems that must be overcome. For example, converting the shoulder to a general-purpose lane will impede incident management activities. The intermittent use of the shoulder for general-purpose travel may increase crashes if drivers fail to obey the overhead instructions regarding lane use. And it may become unacceptably complicated for motorists to navigate getting on and off the freeway at interchanges.
To address some of the safety and operational concerns, mitigating strategies are available, such as installing emergency turnout areas adjacent to the shoulder travel lane, ensuring appropriate design features, and providing motorist guidance at freeway entry and exit locations. But whether these strategies will be sufficient to acceptably reduce the new safety and operational risks introduced by the proposed concept is unknown at this time.
Advantages Of FEE Lanes
The appeal of the new FEE lane concept can be summarized as follows:
Generally, conversions of HOV lanes to HOT lanes have achieved high approval ratings from the public, with approval after implementation rising significantly relative to ratings prior to implementation. For example, in Minneapolis approval ratings among lower and higher income motorists increased after conversion of HOV lanes to HOT lanes on I-394 in spring 2005.
Past attempts to convert a general-purpose lane into an HOV lane have been unsuccessful because of public opposition. "There have been two main public concerns," says Orski. "First, commuters are concerned they will not be able to use the lanes that had been previously available to them; and second, they are concerned congestion will get worse on the remaining general-purpose lanes." In addition, people also perceive equity as a negative for conversions of general-purpose lanes into HOT lanes. The equity issue may be addressed on HOT lanes by providing improved transit services (more likely to be used by moderate and low-income groups), by offering toll-free service for HOVs, and by providing toll discounts or credits for means-tested low-income commuters, for example, those who qualify for the Federal Earned Income Tax Credit.
The proposed new FEE concept attempts, in addition, to address concerns expressed by Orski. First, the same number of general-purpose lanes will be available as before. Second, in the two known cases where attempts were made to convert freeway general-purpose lanes to HOV lanes, negative public reaction occurred after implementation, because traffic congestion became worse in the remaining general- purpose lanes and motorists stuck there could see the adjacent converted HOV lanes were being underused.
With the proposed FEE concept, however, agencies will actively manage all lanes and ensure that converted FEE lanes are fully utilized. In addition, travel speeds initially will improve significantly on the general-purpose lanes due to the extra lane created and increase in overall capacity of the facility. Therefore, public opposition to general-purpose lane conversions is less likely.
Motorists might appreciate the trip time "insurance" benefit that would be available when they need it. Transit and vanpool users might value the travel time savings. Carpoolers might value the option to travel faster by sharing the cost of the tolls, especially on freeways that do not have HOV lanes. Operating agencies might appreciate not having to deal with traffic flow disruptions and additional expenses for onsite enforcement of vehicle occupancy where existing HOV lanes are converted to FEE lanes, because in this approach compliance is enforced through cameras, transponders, and preregistration of eligible HOVs.
MnPASS Acceptance "Good Idea" By Income
Environmental groups, too, could be brought onboard with a more aggressive pricing option, discussed below, which involves pricing of all existing general-purpose lanes. "They might also appreciate the reduced pressure for widening freeways and initiation of the motoring public to direct user charges that better reflect the true costs of highway use, especially if accompanied by improved transit options," says Michael Replogle, transportation director of the Environmental Defense Fund. Pricing all lanes can keep traffic free flowing on all lanes, reducing emissions while at the same time curbing "induced" traffic, that is, keeping total corridor traffic volume the same as before or lower. If toll revenue is used to improve transit and carpool services, the reduction in the number of vehicle trips that results from mode shifts to carpools, vanpools, and transit will reduce vehicle pollutant and greenhouse gas emissions and likely would result in lower air pollution.
Assuming resolution of the outstanding safety issues, the FEE lane configuration involving a single priced lane in each direction might be considered a baseline. More aggressive approaches could produce additional benefits in terms of travel delay reductions and revenue enhancements. For example, an agency could reconfigure a six-lane freeway so that the lane adjacent to the leftmost lane is also priced, making two FEE lanes and two toll-free, general-purpose lanes. Alternatively, the agency could convert all four restriped lanes into priced lanes, with no free service option (the high-performance highway concept discussed in the May/June 2007 issue of Public Roads).
"Tolling all lanes on the entire freeway network in the Seattle metropolitan area would entail a tolling infrastructure cost of approximately $88 million and pay for itself in about 1 year," says Jack Opiola, principal consultant of Booz and Company for Global ITS, who conducted a 2007 study of tolling for the Seattle freeway system. (See Destination 2030-Taking An Alternative Route.) "Adding AcTM is a natural fit, and its costs could be absorbed in the project and improve not only the financial payback but the safety and general traffic management benefits."
Adds John Halkias, traffic analysis team leader at the Federal Highway Administration (FHWA): "There are many potential advantages with these approaches. They could reduce the amount of weaving that is needed to get into and out of priced lanes, thus making better use of available freeway capacity. Having multiple priced lanes increases the likelihood of sustained free-flow traffic conditions, and utilizes available road space more effectively." With a single lane, queues form behind a slower vehicle and a gap is created in front of it, reducing vehicle throughput.
"A single separated lane, such as an HOV lane, has a capacity penalty [a reduction in capacity] of at least 20 percent," says Pravin Varaiya, a researcher at the University of California, Berkeley, who has conducted extensive research on HOV lane performance in that State. Multiple priced lanes prevent this problem, and agencies could provide better service to more motorists, increasing the overall delay reduction and economic benefits.
With more premium-service capacity available, the system operator would be able to collect more in tolls, increasing resources available to maintain and operate the system, notes FHWA's Darren Timothy, an economist in the Office of Innovative Program Delivery.
"Because new capacity is added for use by peak-period travelers, they might be more willing to pay the relatively small charges needed to finance the extra capacity and operate it," says Director Regina McElroy, FHWA Office of Innovative Program Delivery. "Rather than seeing new tolls as punitive, they might see them as beneficial because of the direct benefits they receive from paying the new tolls."
Replogle cautions, however, that many motorists are likely to see new tolls on existing capacity as punitive and unfair unless they are guaranteed that they will not have to pay if they are delayed by congestion on the managed lanes and unless they are offered new and better transit options. Tolling existing lanes can generate significant new revenue, which can be dedicated to funding new transit services, compliance with corridor-level agreements for environmental and system performance, and subsidies for lower income travelers adversely affected by pricing. "All these elements should be considered to increase public acceptability of tolling existing lanes for high performance," Replogle advises.
More Aggressive Pricing Options for an Existing Six-Lane Freeway
In Singapore, net revenue from the existing congestion pricing program goes toward tax rebates of $100-$200 per year to all vehicle owners, regardless of how much they drive. Another way to address public concerns might be to provide all motorists in a region with a limited quota of toll credits. "Credit-based congestion pricing could address both double taxation and equity concerns," says Kara Kockelman, associate professor in the Department of Civil, Architectural and Environmental Engineering at the University of Texas at Austin.
Although congestion pricing can be an effective way to improve the efficiency of the highway system, moving forward with this approach will require that creative solutions be explored and developed with regard to the public acceptance, political, safety, and operational challenges. New technologies and approaches such as AcTM could assist in addressing the challenges. Pricing projects that incorporate AcTM will begin operating in 2010 in Minneapolis and Seattle. They will provide valuable lessons about the feasibility and desirability of such approaches.
The Washington State Department of Transportation (WSDOT) sees a natural fit between AcTM and pricing, whether it is the pricing of single lanes or an entire roadway. WSDOT plans to deliver AcTM in conjunction with planned tolling of the Highway 520 floating bridge in Seattle. The department also will evaluate the potential benefits of combining a priced dual express lane system with AcTM on I-405.
"Over our 30-plus years of freeway management efforts, we have seen firsthand the benefits of lane and roadway management strategies," says WSDOT State Traffic Engineer Ted Trepanier. "We foresee many ways we could apply active traffic management as well as pricing to individual lanes or entire roadways to benefit our customers. What we're considering on Highway 520 and I-405 are just two examples."
The freeway pricing concepts presented here could prove to be a promising solution to metropolitan highway congestion, while at the same time providing a stream of toll revenues that could be leveraged to pay for the concepts' own implementation. However, more detailed assessments of the opportunities and options in specific metropolitan areas are needed. FHWA encourages such studies through its Value Pricing Pilot (VPP) program. FHWA researchers hope this article will stimulate further exploration and discussion of the concepts, and generate other ideas. FHWA also encourages studies to test public acceptability of the concepts in focus groups of transportation users and stakeholders, in order to develop features that address public concerns.
The author thanks WSDOT's Ted Trepanier, Patty Rubstello, and Mark Bandy for their valuable contributions to this article. However, the author alone is responsible for the views expressed and any errors or omissions.
Patrick DeCorla-Souza, AICP, is manager of tolling and pricing programs in the Office of Innovative Program Delivery at FHWA in Washington, DC. He has been involved in managing the VPP program since 1999 and works with public and private sector partners nationwide to implement innovative road-pricing strategies. He cochairs the Transportation Research Board's Congestion Pricing Committee and has master's degrees in transportation planning and civil engineering.
For more information, contact Patrick DeCorla-Souza at 202-366-4076 or firstname.lastname@example.org.