Understanding the Contributions of Operations, Technology, and Design to Meeting Highway Capacity Needs (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

37 C h a p t e r 3 This chapter identifies technologies and non-lane-widening treatments that improve network capacity and reduce the breakdown probability on arterial and freeway facilities and networks. The first section focuses on technologies, both emerging and visionary, that have the ability to positively impact overall network operations. The second section focuses on treatments, some of which incorporate the technologies described in the first section, and organizes the treatments on the basis of their applicability to facility type (arterial, free- way, or both) and their potential for increasing capacity and reducing the probability of breakdown. The research team prioritized both technologies and treat- ments based on an extensive review of past and emerging research as well as discussions with experts in the field. From this prioritization, 25 unique treatments were identified for testing and application in the enhanced network model described in Chapter 2. Key Findings and Conclusions Results from an investigation of technologies that have the potential for widespread implementation in both the near term (less than 10 years) and long term (beyond 10 years) were identified and evaluated on the basis of their potential to improve capacity and throughput, ease of implementation, and timetable for implementation. Table 3.1 summarizes the top 10 most promising technologies. Following a similar approach, a set of 25 non-lane- widening treatments were identified that offer agencies the ability to improve the operational efficiency of their road- way networks. These treatments improve network opera- tions by increasing base capacity, reducing the probability of breakdown, and/or shifting demand to underutilized links on the network. Each of these treatments was consid- ered for testing by using the enhanced operational model described in Chapter 2. Table 3.2 summarizes the 25 identi- fied treatments. technologies affecting traffic Operations Innovation in the transportation sector has provided metro- politan regions with an overwhelming array of possibilities to improve traffic operations and add capacity without construct- ing additional lanes. Through a scan of emerging practices and discussions with experts in the field, technologies were identified that have the potential for widespread implemen- tation within the next 10 years, as well as promising technolo- gies with slightly longer implementation timetables, to reflect the potential for technologies to build on one another with proper planning and foresight. Communications Matrix Two important elements to consider when evaluating tech- nologies are the entities engaged in the transfer of data and the type of data being transferred. The key to maximizing traffic performance, in the form of increased or more reliable throughput, lies in finding ways to allow these entities to inter- act more effectively and efficiently. As an example, consider an individual vehicle as an entity in this framework. The poten- tial for improvement is seemingly endless but becomes much clearer when broken down by the interacting agent. Vehicle- to-vehicle communication—where data concerning speed, congestion, incidents, road conditions, and other factors flow freely between vehicles—may certainly have a significant effect on network performance, but will not likely be commonplace within the next 10 years. Alternatively, localized control devices may serve as the interacting agent, obtaining data concerning speed, turning movements, volume, and the like through sensors or short-range communication to dynami- cally adjust signal timing, ramp metering, and lane availabil- ity. Such interactions may be reached in the next 10 years, but only with a highly organized, unified effort to place the neces- sary systems in the appropriate locations. As a third option, a Operational Technologies and Treatments

38 centralized network management system could also act as the recipient and processor of data sent from individual vehicles. This scenario would require a management system capable of effectively interpreting and distributing the appropriate data, as well as a long-distance communication channel such as a satellite-based system or high-capacity wireless network. The Communications Matrix shown in Table 3.3 provides a means for organizing such emerging technologies based on the source of the data, the intended recipient, the type of data transferred, and the necessary communication channel. Clearly such a premise requires broad, yet distinct, unambiguous, and easily identifiable categories. Four types of sources/recipients seem to emerge: • Vehicle. All data transmitted to or originating from vehicles or their passengers. • Infrastructure. Devices that collect or transmit informa- tional data about the vehicles (e.g., VMS, tag readers) or the infrastructure itself (e.g., ice and snow conditions). • Local control. All local control devices such as signals and ramp meters. • Network control. All control and management systems at the corridor or regional level. When placed in a four-by-four format providing 16 indi- vidual directional categories, the interplay between these entities allows for a variety of ideas while still maintaining accessibility. The source of data, through either direct collec- tion or data synthesis, is identified as the “From” category and is listed across the top. The recipient of that data is listed down the left side of the matrix, labeled “To,” and each indi- vidual component is identified with a directional code. For example, vehicle-to-vehicle communication is denoted by a VV numbered code, network-to-vehicle by an NV, and so forth. The matrix shown in Table 3.3 distinguishes among indi- vidual technologies that share similar objectives but have dif- ferent control methods. For example, dynamically responsive traffic signals could receive information from local control or network control. A localized system bypasses the need for a central agency but necessitates software capable of commu- nicating with nearby signals to accommodate real-time changes in volume, transit priority adjustments, emergency vehicles, and the like. A system based on network control has greater power for platoon formation over long distances, as well as the ability to integrate oversaturation controls but also requires the installation of sensor, software, and communica- tion devices on all control devices within the region of interest. Table 3.1. Top 10 Promising Technologies with High Potential for Improving Capacity and Throughput Potential for Immediate Widespread Implementation Potential for Widespread Implementation in Major Metropolitan Areas Within 10 Years Implementation Likely in 10 or More Years Signal coordination Plus lane Automated vehicle and highway system Electronic toll collection Flow management Reversible-lane control Capacity assignment decisions Network-level ramp metering Route guidance Control coordination Table 3.2. Selected Non-Lane-Widening Treatments to Improve Capacity Freeway Arterial Both HOV lanes Signal retiming Narrow lanes Ramp metering Signal coordination Reversible lanes Ramp closures Adaptive signals Variable lanes Congestion pricing Queue management Truck-only lanes Pricing by distance Raised medians Truck restrictions HOT lanes Access points Pretrip information Eliminate weaving sections Right/left turn channelization In-vehicle info Frontage roads Alternate left turn treatments VMS/DMS Interchange modifications

39 Table 3.3. Communications Matrix Technology Matrix To From Vehicle Infrastructure Local Control Network Control Vehicle VV1 Receive Nearby Vehicle Decisions IV1 Infrastructure Status LV1 In-Vehicle Display of Signals NV1 Route Guidance VV2 Transmit Subject Vehicle Decisions IV2 In-Car Speed Limit Adjustment Display LV2 Reversible Lane Control NV2 Network Highway Advisory Radio VV3 Collaborative All-Way Stop Control IV3 Local Conditions via Advisory Radio LV3 In-Vehicle Lane Assignment Display NV3 511 Assistance VV4 Collaborative Gap Acceptance IV4 Smart Work Zone Navigation LV4 NV4 External Vehicle Speed Control VV5 Adaptive Cruise Control IV5 Automated Parking Enforcement LV5 NV5 VV6 Collaborative Driving System IV6 Intelligent Vehicle and Highway System LV6 NV6 VV7 IV7 LV7 NV7 VV8 IV8 LV8 NV8 Infrastructure VI1 Vehicle Dynamics Info II1 Advance Incident Detection/Warning LI1 Dynamic Advance Warning Signals NI1 Capacity Assign- ment Decisions VI2 Weather/Road Conditions II2 Congestion Detection/ Warning LI2 NI2 Weather Forecasts VI3 Electronic Toll Collection II3 LI3 NI3 Dynamic Congestion Tolling VI4 II4 LI4 NI4 VI5 II5 LI5 NI5 VI6 II6 LI6 NI6 VI7 II7 LI7 NI7 VI8 II8 LI8 NI8 Local Control VL1 Approach Trajectory Input IL1 Intersection Conditions LL1 Signal Coordination NL1 Control Coordination VL2 Turning Movement Options IL2 Dynamic Speed Limits LL2 NL2 Oversaturated Control VL3 Lane Use Options IL3 Blocked Lane Information LL3 NL3 Network-Level Ramp Metering VL4 IL4 Ramp Meter Override LL4 NL4 Flow Management VL5 IL5 Sensor Status LL5 NL5 Plus Lane VL6 IL6 LL6 NL6 Algorithms for Speed Adjustment VL7 IL7 LL7 NL7 VL8 IL8 LL8 NL8 Network Control VN1 Probe Information (AVI, AVL) IN1 Volume Monitoring LN1 Performance Capabilities NN1 Regional Handoffs/ Coordination VN2 Desired Paths IN2 Congestion Monitoring LN2 Local Control System Status NN2 VN3 Desired Arrival Times, etc. IN3 Speed Monitoring LN3 NN3 VN4 Performance Limitations IN4 Unmanned Aerial Vehicles LN4 NN4 VN5 IN5 LN5 NN5 VN6 IN6 LN6 NN6 VN7 IN7 LN7 NN7 VN8 IN8 LN8 NN8

40 As a result, two technologies with similar objectives actu- ally have very different pathways for implementation, as well as very dissimilar methods of combining with other technologies. The following provides a summary of the technologies listed in Table 3.3 and sorted by each of the four origin enti- ties discussed below. Vehicle To Vehicle These include all forms of direct communication between vehi- cles. Adaptive Cruise Control (VV5) is a basic form of this strat- egy in which vehicles automatically adjust their speeds based on distance and approach speed data from other vehicles. As most forms of VV communication require sensors and onboard display devices in both the sending and receiving vehicle, most of these ideas will not be widely available within the next 10 years. However, they are all likely to have dramatic effects on traffic safety and sustainable flow, such as VV4 Collaborative Gap Acceptance, which provides lane availability information among vehicles to ease merging, or information regarding the safety of turning in front of oncoming vehicles. Examples of these technologies are the ongoing VII program in the United States and the ubiquitous transportation (u-T) networks research program in South Korea. A schematic of the u-T system is depicted in Figure 3.1. To infrasTrucTure In the absence of a GPS-based system, in which vehicles com- municate directly with a central management system, vehicle- to-infrastructure communications such as the one tested by Demers and List (1) provide second-hand information to other drivers concerning localized conditions. VI1 Vehicle Dynam- ics and VI2 Weather/Road Conditions allow vehicles to pass along information regarding speed, rapid braking, or pave- ment conditions to roadside systems such as VMS to inform digital warnings. VI3 Electronic Toll Collection is a more imme- diate representation of this category, which certainly has the potential to improve traffic operation, especially if progressed to the point of open-road tolling. To local conTrol A basic form of this type of technology is outlined by Gradinescu et al. (2). Vehicles relay speed data to signals, thereby expand- ing the range for which signals can identify approaching vehicles as well as bypass the need for sensors. VL2 Turning Movement Options and VL3 Lane Use Options are both more advanced examples, in which vehicles relay their turn- ing intentions or lane preferences to the signal for dynamic adjustments. To neTwork conTrol VN1 Probe Information and VN2 Desired Paths are examples of emerging technology with probe vehicles and GPS units aiding network management processes by providing both real-time travel information and route intentions for travel forecasts. As network management systems grow and develop, other useful information may emerge such as VN4 Perfor- mance Limitations, which relays speed, stopping ability, and traction information to the central system. Infrastructure To Vehicle IV3 Advisory Radio provides a common example of this type of technology in use today. IV4 Smart Work Zone Navigation is beginning to emerge in the form of variable message signs, but throughput and safety may improve greatly as systems develop that allow work zone information to pass directly to vehicles through in-dash display, aiding speed management and merging. IV6 Intelligent Vehicle and Highway System is possibly the most obvious example of how technology will eventually completely overhaul transportation systems by allowing vehicles to coordinate with roadside sensors and computers to enter into an automatic mode, increase speed, and greatly reduce following distances. To infrasTrucTure Similar to vehicle-to-infrastructure technologies, these types of technologies simply have a slightly altered path- way. Sensors, rather than vehicles, collect volume, speed, or incident data and pass this information on to roadside digi- tal signs. To local conTrol This type of technology exists today in the form of detectors used at signalized intersections, but technological improve- ments will add new capabilities to sensor-to-control devices. IL1 Intersection Conditions and IL2 Dynamic Speed Limits (Figure 3.2), for example, would collect weather information and pavement conditions to inform signal timing adjustments Figure 3.1. Vehicle-to-vehicle communication (u-T network, South Korea).

41 or safe speed limits displayed on digital overhead signs based on current weather conditions. IL4 Ramp Meter Override sys- tems will also prove necessary within-ramp metering schemes to prevent arterial backups. To neTwork conTrol In addition to information collected directly from vehi- cles, pavement and roadside sensors will continue to play important roles in collecting speed and volume informa- tion for use at network management scale. While expensive, satellite-based systems (IN2 Congestion Monitoring) and IN4 Unmanned Aerial Vehicles may also prove valuable tools when used in conjunction with a centralized network management system. Local Control To Vehicle Although these systems require some form of an onboard display system, suggesting a long path toward implementa- tion, LV1 In-Vehicle Display of Signals may have significant safety implications, and LV3 In-Vehicle Lane Assignment Display could potentially have a large enough organizational effect on traffic to gain additional throughput. Reversible-lane controls, while already widespread, can lead to significant capacity improvements, especially when used in conjunction with real-time sensor, probe, or network management data. To infrasTrucTure Schultz (3) outlines the effects of Dynamic Advance Warning Signals, which provide early warnings to vehicles regarding current signal status via roadside or overhead displays. Such tools may become much more useful as signals advance from pre-timed to dynamically adjusted schemes, possibly leading to shortened or eliminated amber and all-red phases. To local conTrol As previously discussed, LL1 Signal Coordination allows for local adjustments to cycle length, splits, and offsets through simple communication with nearby control devices, bypass- ing the need for network-level management. To neTwork conTrol To allow for network-level management of arterials, systems such as LN1 Performance Capabilities and LN2 Local Control System Status will need to be in place to relay the capabilities of the control devices, such as max greens, as well as the cur- rent status of cycle lengths and broken loops. Network Control To Vehicle While NV3 511 Assistance currently serves as a major focus of information distribution regarding current roadway condi- tions, NV1 Route Guidance systems that utilize in-vehicle dis- play of optimal routes will likely emerge as a more powerful tool. Demers and List (1) experimented with a system that used Wi-Fi, a pocket PC, and a synthesized voice to communicate route guidance to the driver. At this point it is unclear whether systems such as NV4 External Vehicle Speed Control, which limit a vehicle’s speeds on the basis of its network location, will come into widespread use before an automated highway sys- tem eliminates the need for such an unpopular system. To infrasTrucTure The use of VMS to route traffic on the basis of current con- gestion (NI1 Capacity Assignment Decisions) or display area weather forecasts (NI2 Weather Forecasts) is emerging as a vital tool, especially during the long adjustment window needed for widespread adoption of in-vehicle display units (Figure 3.3). Similarly, NI3 Dynamic Congestion Tolling provides a method for which network management systems can adjust tolls, creating greater incentive to follow network-suggested routes. Figure 3.2. Variable speed limit (VSL) control in Stockholm, Sweden. Figure 3.3. VMS queue-based routing in the Netherlands.

42 To local conTrol Network control over local control devices will likely emerge in a number of different ways, each intended to use real-time information processed at a central network agency to dynam- ically adjust traffic flow. NL3 Network-Level Ramp Metering and NL2 Oversaturated Control, where a network system completely overrides all signals in a region to flush an arterial network, are both powerful examples of this practice. NL5 Plus Lane is a more specific example of such a practice. Widely used in the Netherlands, as outlined by the International Tech- nology Scanning Program (4), a Network Management Sys- tem takes control over a freeway during congested conditions by opening the narrower left shoulder to traffic and lowering the speed limit across all lanes as depicted in Figure 3.4. To neTwork conTrol As network management systems expand their scope, there will likely be a need to break larger metropolitan networks into smaller regions with open sharing of information. Addi- tionally, network-to-network communication may even be useful over greater distances to inform adjacent networks of approaching trucks or high volumes. Relevance Assessment The research team developed a method to rank and identify technologies that are believed to have the most immediate potential to provide improvements in traffic operations. Four categories of criteria were developed for the ranking: • Timetable. A 1 through 3 grading scale was used on the basis of the following criteria with 1 representing the lowest score and 3 the highest: (a) technologies anticipated to exist in test-bed environments in the next 10 years, (b) technolo- gies likely to see implementation in progressive metropoli- tan regions in the next 10 years, and (c) technologies likely to have widespread implementation in the next 10 years. • Ease of implementation. A 1 through 3 grading scale was used on the basis of the following criteria, with 1 represent- ing the lowest score and 3 the highest: (a) technologies with a large number of barriers, (b) technologies with a few barriers, and (c) technologies with relatively few barri- ers to implementation. • Capacity. A 1 through 5 grading scale was used to stratify each technology on the basis of its expected impact on the ability of the system to process vehicles, with a 1 suggesting minimal impact, ranging up to a very significant improve- ment denoted by a 5. • Throughput. A 1 through 5 grading scale was used to stratify each technology on the basis of the number of vehicles pro- cessed per unit time, with 1 suggesting minimal impact, ranging up to a very significant improvement denoted by a 5. The research team applied individual knowledge, experi- ence, and projections concerning each technology. While such a system is subjective in nature, it allows for a blending of ideas based on experienced judgment. This approach was further refined during a number of discussions in which each team member defended his or her rankings and assumptions. Table 3.4 presents the compiled rankings and averages. For the Timetable and Ease of Implementation categories, the averaged results were rounded to the nearest whole number, creating a final 1 through 3 ranking for each technology in each category. The Capacity and Throughput scores were also averaged, but the decimals were retained. The mean of these two averages is shown in the “Average Score” column. Table 3.4 is sorted first by Timetable, then by Ease of Implementation, and finally by Average Score. The widely available and fairly easy-to-implement technolo- gies rise to the top of Table 3.4, while those that are farthest off and most difficult to implement fall to the bottom. To balance near-term accessibility with significant long-term benefits, three cutoff scores were instituted (3.0, 3.5, and 4.0), increasing in value as the timetable for implementation increases. Those technologies that are available (3) and fairly easy to implement (2 or 3) needed to score higher than average (3 or higher) in order to achieve relevance for the purpose of this discussion. Five of the 11 technologies in the top category topped this average score threshold: • Signal Coordination (LL1); • Electronic Toll Collection (VI3); • Reversible Lane Control (LV2); • Network-Level Ramp Metering (NL3); and • Control Coordination (NL1). Figure 3.4. Implementation of plus lane (closed condition) in the Netherlands. (text continued on page 46)

43 Table 3.4. Ranked Technology Options Ref Technology Details Timetable (1  test bed; 2  selective; 3  widespread) Ease of Implementation (1  low; 2  medium; 3  high) Average Score 1 Signal Coordination Local adjustments to cycle length, splits, offsets, without network-level control or guidance 3 3 3.21 2 Electronic Toll Collection Toll collection from RFID or tag readers 3 3 3.14 3 Network Highway Advisory Radio Networkwide radio broadcast of conditions/problem areas/detours 3 3 2.71 4 Volume Monitoring Pavement sensors relay real-time highway volume information to network management system 3 3 2.57 5 Congestion Detection/ Warning Use of detection devices and VMS to automatically display congestion warnings/ delay 3 3 2.57 6 Local Conditions via Advisory Radio Localized radio broadcast weather/congestion/emergency conditions 3 3 2.07 7 Reversible Lane Control Reversible lane control and indications. 3 2 4.14 8 Network-Level Ramp Metering Network coordination of ramp meters for optimal highway flow 3 2 3.93 9 Control Coordination Network-level adjustments for signal control 3 2 3.50 10 Local Control System Status Ability to function, broken loops, current cycle length, splits, offsets, patterns in these same parameters over recent time, flash, emergency preemptions if they have occurred, and so forth 3 2 2.86 11 Ramp Meter Override Queue detection allows for ramp meter override to prevent intersection backup 3 2 2.57 12 Plus Lane Use of overhead signs to reduce speed limit and open left shoulder of a freeway to increase capacity during high volumes 2 2 4.07 13 Flow Management Use of dynamic devices including overhead lights, in-pavement lighted lane dividers, or automatic movable barriers in order to maximize flow by opening or closing shoulder lanes or express lanes as needed for network optimization 2 2 3.86 14 Capacity Assignment Decisions Use of VMS to route traffic based on network information (around incidents, alternate routes to avoid congestions and so forth). 2 2 3.57 15 Route Guidance In-vehicle display of suggested route based on current/projected conditions 2 2 3.50 16 Oversaturated Control Signal timing to flush arterial networks. Ramp meter closures. Signal timing plan changes to ameliorate oversaturation. Changes in lane use to provide surge capacity in specific directions on specific facilities. 2 2 3.43 17 Dynamic Congestion Tolling 2 2 3.14 18 Advance Incident Detection/ Warning Use of detection devices and VMS to automatically display incident warnings 2 2 3.00 19 Blocked Lane Information Infrastructure senses breakdowns, or degradations in lane condition and relays this information to local control 2 2 2.79 (continued on next page)

44Table 3.4. Ranked Technology Options (continued) Ref Technology Details Timetable (1  test bed; 2  selective; 3  widespread) Ease of Implementation (1  low; 2  medium; 3  high) Average Score 20 511 Assistance Interactive route guidance info via cell phone 2 2 2.64 21 Performance Capabilities Performance capabilities of the local control: cycle length, max greens, storage lane lengths, where variable, spillback limitations, to the extent that they vary, sensed throughput capabilities, such as variations in saturation flow rates due to sun glare, local work zones, intense parking, and so forth. 2 2 2.64 22 Automated Parking Enforcement Example from South Korea: Sensors and digital video collect tag info from illegally parked vehicles and automatically broadcast warning to owner over speakers/cell phone/Internet, and issue tickets by mail. 2 2 2.57 23 Probe Information (AVI, AVL) Vehicle tracking data for network use 2 2 2.42 24 Dynamic Speed Limits Speed limit adjustments made based on pavement conditions and weather 2 2 2.14 25 Speed Monitoring Roadside sensors relay real-time highway speed information to network management system 2 2 2.14 26 Weather Forecasts Network weather station provides localized forecasts to VMS 2 2 2.14 27 Sensor Status Open channel communication between sensors and local control (faulty loop, stalled vehicle) to prevent poor control due to inaccurate information 2 2 2.00 28 Dynamic Advance Warning Signals Overhead warning of signal status in advance, useful for high-speed intersections 2 2 1.71 29 Adaptive Cruise Control Automatic cruise control which maintains a minimum following distance 2 1 3.14 30 Algorithms for Speed Adjustment Network controlled algorithms for localized speed adjustment based on network conditions 2 1 2.50 31 Lane Use Options Lane use options that the vehicles see, for example: need to be in the RH lane, need to be in the LH lane, indifferent about the through lanes, based on immediate and upcoming turning movements; also dimensional restrictions, such as need to be in the center lane, or restriction limitations, such as trucks have to be in the center lane. 1 2 2.93 32 Desired Arrival Times Vehicle transmits desired arrival time at a specific location for feedback regarding best available route. 1 2 2.86 33 Smart Work Zone Navigation Virtual cones, work zone navigation assistance—cone locations, trajectories, paths to follow, smoother merges 1 2 2.50 34 Intersection Conditions Local control parameters adjust for ice, snow, sun glare, and so forth. 1 2 1.93 35 Weather/Road Conditions Adjust infrastructure conditions (e.g., salt application) in response to vehicle behavior information 1 2 1.64 36 Intelligent Vehicle and Highway System Driver inputs their destination and hands off vehicle control to an onboard com- puter when entering a freeway, which coordinates with other vehicles and roadside infrastructure. The intelligent system then optimizes lane usage and spacing with other intelligent cars, forming platoons and maximizing capacity. 1 1 4.21 (continued on next page)

45 Table 3.4. Ranked Technology Options (continued) Ref Technology Details Timetable (1  test bed; 2  selective; 3  widespread) Ease of Implementation (1  low; 2  medium; 3  high) Average Score 37 Collaborative Driving System Automated driving system utilizing platoon formation and coordination with other vehicles 1 1 3.64 38 Collaborative Gap Acceptance Yield or stop-controlled locations or lane changing on freeways or arterials, what are the gaps that can be collaboratively accepted or must be rejected, vehicles working together to accommodate needed lane changes; the idea is akin to forced merges, but a lot more polite. 1 1 3.00 39 Desired Paths Vehicle transmits expected route to network management system. 1 1 2.93 40 In-Vehicle Lane Assignment Display Local control provides vehicle lane assignment directly through in-vehicle display. 1 1 2.79 41 Congestion Monitoring Satellite-based sensors relay imagery and traffic monitoring data to network management system. 1 1 2.71 42 Infrastructure Status Intelligent speed adjustment and settings for braking, acceleration, and so forth. 1 1 2.43 43 Approach Trajectory Input Vehicle relays approach speed and vehicle type (emergency, public transit, HOV, etc.) to signal. 1 1 2.36 44 Receive Nearby Vehicle Decisions In-car display/broadcast of nearby vehicle movements 1 1 2.36 45 Collaborative All-Way Stop Control Vehicle communication and preference relay at all-way stop. 1 1 2.29 46 Turning Movement Options Vehicle relays turning movement intention to signal. 1 1 2.25 47 In-Car Speed Limit Adjustment Display Vehicle receives variable speed limit information for in-dash display. 1 1 2.21 48 Vehicle Dynamics Info Informs speed limit adjustments, heads-up display messages, and so forth. 1 1 2.07 49 External Vehicle Speed Control (EVSC) Controls vehicle speed based on network speed limit maps and onboard GPS/ EVSC system 1 1 2.00 50 Unmanned Aerial Vehicles Unmanned aircraft for continuous traffic surveillance 1 1 1.93 51 Regional Handoffs/ Coordination Air Traffic Control handoff-type thoughts 1 1 1.83 52 In-Vehicle Display of Signals Allows for in-car display of signal/ramp meter status at time of arrival. 1 1 1.79 53 Performance Limitations Acceleration, deceleration, stopping distance, adhesion (snow, ice), safe following distances as observed by the vehicle control system. 1 1 1.71 54 Transmit Subject Vehicle Decisions Signaling/brake light info transmitted directly to nearby vehicles 1 1 1.64

46 The next 17 technologies will take slightly longer to achieve widespread availability (2) and have a few, but not an over- whelming number of, barriers to implementation (2). Of these, 4 technologies score higher than 3.5: • Plus Lane (NL5); • Flow Management (NL4); • Capacity Assignment Decisions (NI1); and • Route Guidance (NV1). The remaining technologies contain a 1 for either time- table or ease of implementation, suggesting that they are likely too distant or too difficult to implement for the pur- pose of this discussion. Intelligent Vehicle and Highway Sys- tem (IV6), however, scores the highest of any technology considered. Inventory of Network Operations treatments Treatments refer to the actions and applications that have the potential to improve sustainable service rates and reduce the probability of breakdown along freeway and arterial facilities and networks. Nearly 100 treatments were identi- fied through a comprehensive literature search and discus- sions with technical experts in the field. In some cases the treatments incorporate technologies identified in the previ- ous section. The team ranked the effectiveness of each treatment with respect to the following three criteria: • Effect on peak-hour congestion. The potential for the treat- ment to reduce peak-hour recurring congestion by increas- ing capacity and/or reducing the probability of breakdown. • Purview of decision makers. Degree to which agency deci- sion makers have the ability and institutional authority to implement the treatment. • Ease of implementation. Ease of implementing the treat- ment considering economic, social, political, and environ- mental costs. Based on the results of the ranking process, and following a consolidation of treatments into broader categories, a set of 10 broad categories emerged as effective, viable, and ready for implementation. Table 3.5 presents these categories and char- acterizes them according to their individual ability to increase capacity and decrease the probability of breakdown on free- ways and arterials. The following sections provide a brief description of each treatment as well as a description of the capacity-enhancing effects, known applications, and implementation needs. Lane Treatments Lane treatments result in an added or dedicated travel lane for a directional movement of traffic and/or a specific vehicle/ user type in the current paved section of roadway. Their objec- tive is to improve the vehicle-moving capacity for peak direc- tional movements during congested periods of the day. Lane treatments can also be applied to increase the people-moving capacity of the facility and reduce travel demand in the case of bus-only or HOV lanes. Lane treatments apply to both free- way and arterial facilities and can be implemented on a static or dynamic basis. Lane treatments represent the most com- mon class of treatments for recurring bottlenecks. Common examples of lane treatments include • Narrow lanes/use of shoulder lanes; • Reversible lanes for arterials; • HOV lanes on freeways; • Variable lane controls at a signalized intersection; and • On-street parking restrictions during peak periods. Narrow lanes refer to adding a travel lane by re-striping a roadway with narrower lanes and/or converting part or all of the shoulder to a travel lane (also referred to as a “plus” lane, as described in Table 3.2). Reversible lanes are used on arterial roadways, freeways, and bridges/tunnels to increase capacity for facilities that have directional peak traffic flows. Most reversible-lane applica- tions on freeways are implemented by constructing a sepa- rated set of lanes in the center of the freeway with gate controls on both ends. For undivided facilities, a movable barrier can be applied to physically separate opposing directions of traf- fic flow. Reversible lanes on arterials are typically implemented by using DOWNWARD GREEN ARROW and RED X lane- use control signs as described in the Manual on Uniform Traf- fic Control Devices (5). Implementation issues include driver awareness/education, enforcement, safety (potential for head- on collisions), maintenance and operations, and accommoda- tion of left turn movements (6). An HOV lane is reserved for the use of carpools, vanpools, and buses; motorcycles can usually use them as well. Most HOV lanes are applied to freeway facilities next to unrestricted gen- eral purpose lanes, but some are also used on arterial roadways. HOV lanes are intended to increase the person-moving capac- ity of a corridor by offering incentives for improvements in travel time and reliability. An inventory of existing and planned HOV facilities is provided through FHWA’s Office of Opera- tions website for HOV facilities. Other lane treatments include converting a closely spaced on-off ramp sequence to a weaving section by extending the acceleration and deceleration lanes into a full auxiliary lane, (continued from page 42)

47 Table 3.5. Summary of Treatments for Achieving Improvements in Network Operations Treatment Freeway Arterial Increases Capacity Decreases Probability of Breakdown Increases Capacity Decreases Probability of Breakdown Operational Treatments Lane Treatments - Narrow lanes - Reversible lanes - HOV lanes - Variable lanes - On-street parking restrictions • • • • Signal Timing - Signal retiming - Adaptive traffic control - Queue management - Transit/truck signal priority • • Traffic Demand Metering - Ramp metering - Mainline metering - Ramp closures - Arterial demand metering • • Congestion Pricing - Pre-set pricing - Dynamic pricing - Distance/vehicle class tolls - High-occupancy tolls - Central area pricing • • Traveler Information - Pretrip information - In-vehicle information - Roadside messages - GPS navigation devices • • Variable Speed Limits • • Design Treatments Access Management - Raised medians - Access consolidation/relocation - Right turn channelization - Frontage roads • • Geometric Design Treatments - Flyovers - Improving weaving sections - Alternate left turn treatments - Interchange modifications - Alignment changes • • • • Truck-Related Treatments Truck/Heavy Vehicle Treatments - Truck-only lanes - Truck restrictions/prohibitions - Truck climbing lanes • • • •

48 adding lanes at off-ramps to mitigate the impact of a down- stream signal on the surface street of an interchange, and adding temporary (median) lanes at weaving sections during peak periods. Variable lanes at an intersection refer to the use of variable lane-use control signs that change the assignment of turning movements to accommodate variations in traffic flow. Con- versely, many jurisdictions restrict left turn movements at intersections during peak periods through static or variable signing. While variable lanes are currently applied on a time- of-day basis, variable lanes could be applied dynamically with the proper technology and driver education/enforcement in place. The use of variable lanes to add turn lanes requires adequate turning radii, presence of a sufficient number of receiving lanes, variable-mode signal phasing, and advance warning signs. On-street parking restrictions are often applied on urban roadways to provide additional through-capacity during peak commute periods and to preserve parking for local uses dur- ing off-peak periods. Enforcing the parking restrictions is a key challenge given that the potential capacity associated with the parking lane may not be achieved if one or more vehicles remain parked in violation of the restriction. Signal Timing Signal retiming is a process that seeks to optimize the control- ler’s response to roadway user demand by implementing or modifying signal timing parameters (i.e., phase splits, cycle length, and offset), phasing sequences, and control strategies. Signal retiming can be carried out for an individual inter- section, an arterial corridor, or an entire network. Effective signal retiming can increase capacity and reduce signal delay, which leads to lower travel times, improved reliability, and reduced driver frustration. Significant benefits can be achieved for intersections that have experienced changes in traffic flows and arrival patterns since the timing plans were last updated. Adaptive traffic control systems (ATCS) use algorithms and system detectors to perform real-time optimization of traffic signals based on current or anticipated future traffic conditions. The adaptive software adjusts signal splits, offsets, phase lengths, and phase sequences to achieve a defined objective (e.g., mini- mize delay, reduce stops). There are five types of adaptive traffic control systems in use today. Two of the first ATCS systems developed are SCATS and SCOOT. SCATS (Sydney Coordi- nated Adaptive Traffic System) was developed in Australia in the early 1970s and SCOOT (Split Cycle Offset Optimization Technique) was developed in the United Kingdom a few years later. Recently, in the United States, other adaptive control sys- tems have been implemented and/or are in testing, including RHODES (Real-Time Hierarchical Optimized Distributed and Effective System), OPAC (Optimization Policies for Adaptive Control), and FHWA’s ACS-Lite. Queue management is a signal timing technique for over- saturated arterials that seeks to minimize queue spillback within turn lanes or between links. A range of queue manage- ment techniques have been developed and applied around the world. Such techniques include zero offsets, reverse progression, gating, and metering (7–9), and diversion (10). These techniques are particularly critical at closely spaced intersections with limited queuing space, where queue spill- back can block entries into critical intersections and cause a major reduction in arterial throughput (11). Transit/truck signal priority gives special treatment to par- ticular modes such as transit vehicles or trucks at signalized intersections. It does this by either extending the green phase or truncating the red phase for the approaching vehicle. It is unlike signal preemption in that it does not disrupt signal progression. The primary benefit of transit/truck signal pri- ority is improved schedule reliability for transit vehicles and improved capacity/safety for the coincident traffic stream. Transit signal priority has also been shown to reduce travel time for transit vehicles. Transit signal priority does not have a significant effect on improving the vehicle-moving capacity of an arterial, although mainline movements typically benefit because green time increases when a priority call is placed. However, truck signal priority can have a significant effect on the capacity of the affected arterial movement. Traffic Demand Metering Traffic demand metering has useful applications for both freeways and, to a lesser extent, arterial networks. Demand- metering techniques are typically based on the goal of reduc- ing the probability of breakdown of a freeway or major roadway by controlling the rate and location of additional new demand (e.g., from on-ramps and toll plazas). The metered traffic is allowed to enter the freeway or major road at a rate that is compatible with continuous or “sustained service flow” on the mainline. When appropriately applied, demand meter- ing can increase the capacity of freeway and major road sec- tions, and can contribute to the goal of this research, that is the maintenance of a sustained service rate (SSR). Metering may accrue other benefits, including those related to safety and the environment. Metering can also improve travel time reliability. The three principal control methods for demand metering include local pre-timed, local-traffic responsive, and system- wide traffic responsive. These three approaches depend on the availability of local and system sensors as well as the abil- ity of a local site (e.g., one on-ramp) to communicate with nearby on-ramps or an entire system of ramps.

49 One special application of ramp metering is the closure of one or more on-ramps during congested peak periods. In effect, this action does not increase demand on the down- stream freeway section, but does require a high level of user information and signing to divert vehicles from the on-ramp to the adjacent arterial system. The use of demand-metering techniques on a network of arterial streets is typically a more subtle form of freeway ramp metering. Traffic signal timing replaces ramp meters as the key element. Timing can be made more restrictive upstream of a bottleneck to reduce the rate of new demand that reaches the bottleneck. Thus, the delay and diversion activity is increased at the less congested upstream location while less demand is placed on the bottleneck. The implementation of any demand-metering treatment requires a host of supporting actions, including traffic monitor- ing, communications, and control algorithms, typically based on a combination of historical and real-time information. Congestion Pricing Congestion pricing, also known as value pricing, implements a special type of toll to reduce traffic volume during particular times of congestion or in particular areas of congestion. Con- gestion pricing does not increase capacity but rather reduces the chance of breakdown during the most critical hours. The toll changes driving behavior by serving as an incentive for drivers to travel at different times, to find alternate routes, or to choose other methods of travel (e.g., mass transit, carpool- ing). It is meant to encourage drivers to be more conscientious and mindful of their driving habits. There are several different ways to implement pricing. Pre- set pricing involves fluctuating tolls depending on the time of day, even if traffic flow is not congested. Conversely, if the pricing is a function of traffic flow, then the toll will fluctuate depending on the amount of congestion. There are different types of pricing: distance/vehicular classification types, open road types, and closed road types (12). Distance/Vehicle Class tolls include tolls for driving a cer- tain distance or for driving a truck. Open-road tolling involves tolling one way. Two examples of open-road tolling are high- occupancy-toll (HOT) lanes and express lanes. Both types are toll lanes adjacent to non-toll lanes on a roadway that allow drivers to have the opportunity to pay a toll to avoid congestion. New tolling technology can help reduce the complexity and improve the accuracy of the congestion pricing as well as expedite the tolling process. Technology such as electronic tolling helps reduce the delay for paying the tolls. While current experience and research indicate that con- gestion pricing has great potential, the ease of implementa- tion and actual benefit will vary for each application. Political and public support is a key factor for implementation, as is the availability of alternate routes. Without alternate routes, an economic benefit will be achieved but capacity will not improve and significant driver resentment could result for the additional tolls. A concept explored by FHWA to address equity concerns is called FAIR (Fast and Intertwined Regular) lanes (13). FAIR lane pricing creates tolled (express) lanes and non-tolled (gen- eral purpose) lanes on a freeway. When motorists use the gen- eral purpose lanes during rush hour, they would be compensated with credits that could be applied to the express lanes. Traveler Information Traveler information is information that can be provided to the driver that will allow him or her to make a well-informed deci- sion regarding (a) what mode to take, (b) when to depart, and (c) the best route to travel. This information can be provided before and/or during the trip through the Internet, telephones, television/radio, roadside signs, and in-car displays and devices. Many road agencies are implementing ATIS. These systems incorporate close-to-real-time information on roadways col- lected through cameras and traffic reports. The information gathered can be sent out through highway agency Internet sites or via private Internet sites. Dynamic message signs (DMSs) and variable message signs (VMSs)—electronic signs that can be changed to pro- vide current traveler information, as well as alternate route information—are being applied by highway agencies along highly traveled routes to provide updated information to travelers while they are en route. Variable Speed Limits Variable speed limits are applied to freeway sections, primar- ily in metropolitan areas with large traffic volumes and dis- play traffic-actuated speed limits on variable message signs. The primary objective of variable speed limits is to increase road safety and homogenize traffic flow. Variable speed limit systems usually combine speed limit signs with other traffic signs or text messages that are used to display incident or congestion warnings. Variable speed limit systems are often implemented in conjunction with auto- mated speed enforcement. In many applications, variable speed limits are embedded in traffic control systems that also adopt other measures such as lane control, ramp metering, or temporary hard shoulder running. An efficient way to motivate drivers to display adequate speed behavior is to provide the traffic adaptive indication of expected travel times (in minutes) to characteristic points along the freeway (e.g., to well-known large intersections)

50 by VMSs. This application contributes to a more patient behavior in case of congestion and a less nervous style of driving in flowing traffic. Access Management Access management is the “systematic control of the location, spacing, design, and operation of driveways, median open- ings, interchanges, and street connections to a roadway” (14). The intent of access management is to provide access to land development while still maintaining a safe and efficient trans- portation system. The Access Management Manual identifies the following principles for maintaining land-use access and improving the safety and operations of the arterial roadway: • Limit direct access to major roadways. • Locate signals to favor through movements. • Preserve the functional area of intersections and inter- changes. • Limit the number of conflict points. • Separate conflict areas. • Remove turning vehicles from through-traffic lanes. • Use non-traversable medians to manage left turn movements. • Provide a supporting street and circulation system. Raised medians are applied to reduce turning movements and manage access to land uses along a corridor. Implement- ing a full barrier limits the number of interruptions in traffic flow. In addition to a full barrier, a limited access barrier can provide opportunities for drivers to make left turn move- ments where the agencies deem safe and appropriate. There are also different techniques for improving the “mar- gins” on outer edges of freeways or arterials to help improve capacity. First, lanes can be wider near the side of the road to provide more room for cars to maneuver. Another technique is to channelize right turn movements to minimize imped- ances to through movements. The use of frontage roads also creates separation between through and turning/local traffic that, in turn, can improve the capacity and sustainable service rate of the arterial roadway. Geometric Design Improvements Geometric design improvements refer to spot reconstruction or minor geometric widening that can be performed within the existing paved area. They are generally considered low- or moderate-cost improvements that are less significant than a major capital improvement project. They are often alterna- tives to facility widening projects. Flyovers apply to interchange ramps and major through or turn movements at intersections. They are generally consid- ered a spot treatment to address a high-volume movement as opposed to full reconstruction or lane widening of a facility. Flyover ramps can also be applied to at-grade intersections for high-volume left turn movements. A similar concept for urban areas is to depress the major through movement below the grade of the intersection. Flyovers or grade-separated movements add capacity by separating the traffic demand for the subject movement from conflicting flow. Improving weaving sections primarily applies to freeways but can also apply to arterials. The improvement or elimi- nation of weaving sections can be accomplished through changes in striping and lane assignment, use of medians to physically separate traffic flows, reconfiguration of ramps to add/remove movements, and realignment of ramps to increase weaving distance or remove the weaving move- ment. Weaving sections reduce speeds, capacity, and reli- ability (in addition to contributing to safety deficiencies). Improving weaving sections could potentially increase the roadway capacity to that of a basic freeway section or ramp merge/diverge. Alternate left turn treatments for intersections refer to non- conventional intersections that convert left turn movements into other intersection movements in order to reduce the left turn signal phase. Examples include continuous flow inter- sections, jughandle intersections, superstreet intersections, and median U-turns. Alternate left turn treatments increase capac- ity and safety by eliminating one or more left turn phases and allowing more green time for remaining movements. Improved signal progression is generally achieved through a reduced sig- nal phase. Interchange modifications include changes to the inter- change type, ramp configurations, and traffic control of the ramp terminals. An example of modifying an interchange type is converting a full cloverleaf interchange into a partial cloverleaf interchange to eliminate weaving sections. Other interchange modification techniques that could be made to increase freeway capacity include adding lanes on the entry or exit ramps, increasing the storage distance of on-ramps, changing the ramp alignment to reduce or increase travel speeds, and adding signalization or roundabout control at the ramp terminals. Ramp closures or restrictions to one or more vehicle types can be applied on a dynamic, temporary, or per- manent basis to improve freeway performance. Horizontal/vertical alignment changes apply to both free- ways and arterials, and primarily older facilities that were designed and built before modern-day roadway design stan- dards were put in place. Sharp horizontal or vertical curves affect the speed profile of vehicles and, anecdotally, can lead to sudden braking and increase the probability for breakdown.

51 Truck/Heavy Vehicle Restrictions The management of truck traffic can have a positive impact on traffic operations for both freeways and arterial facilities. Traffic operations improvements relating to trucks and to capacity and sustained service rates may require additional factors to be considered such as allowable delivery schedules, weight restrictions on arterial streets, and vertical clearance limits. There are many possible truck operations techniques that relate to increasing capacity on freeways and major arte- rials. The most frequently used management techniques by state departments of transportation (DOTs) surveyed as part of NCHRP Synthesis 314 (15) include • New or improved pavement; • Truck climbing lanes; • Lane restrictions for trucks; • Restriction/prohibition of trucks on specific roads; • Truck parking restrictions/prohibitions; • Improved incident management; • ITS strategies; • Intelligent warning devices; • Weigh-in-motion; • Improved warning signing; • Electronic screening; and • Enhanced enforcement. The major impacts on capacity created by removing trucks from mixed-use lanes are reduced headways, more consistent speeds, and possibly increased SSRs. Selection of Operational Treatments for Testing Consideration Following the comprehensive inventory and review of opera- tional strategies and treatments identified in Table 3.5, a set of 25 strategies was selected for consideration in testing in the operational model developed as part of this project. The 25 strategies were selected on the basis of their ability to affect positive change in network operations. The treatments improve network operations by increasing base capacity, reducing the probability of breakdown, and/or shifting demand to under- utilized links on the network. The 25 strategies that were tested as part of this research project have been summarized in Table 3.2, organized by their application to freeway facilities, arterial facilities, or both. references 1. Demers, A., and G. F. List. Probes as Path Seekers: A New Paradigm. Transportation Research Record: Journal of the Transportation Research Board, No. 1944, Transportation Research Board of the National Academies, Washington, D.C., 2006, pp. 107–114. 2. Gradinescu, V., C. Gorgorin, R. Diaconescu, V. Cristea, and L. Iftode. Adaptive Traffic Lights Using Car-to-Car Communication. Presented at IEEE 65th Vehicular Technology Conference, Dublin, 2007. 3. Schultz, G. G., R. Peterson, D. L. Eggett, and B. C. Giles. Effective- ness of Blank-Out Overhead Dynamic Advance Warning Signals at High-Speed Signalized Intersections. Journal of Transportation Engineering, Vol. 133, No. 10, 2007, pp. 564–571. 4. Mirshahi, M., J. Obenberger, C. A. Fuhs, C. E. Howard, R. A. Krammes, B. T. Kuhn, R. M. Mayhew, M. A. Moore, K. Sahebjam, C. J. Stone, and J. L. Yung. Active Traffic Management: The Next Step in Congestion Management. FHWA-PL-07-012. International Tech- nology Scanning Program, FHWA, 2007. 5. Federal Highway Administration (FHWA). Manual on Uniform Traffic Control Devices. FHWA, U.S. Department of Transporta- tion, 2003. 6. Rodegerdts, L. A., B. L. Nevers, B. Robinson, et al. Signalized Inter- sections: Informational Guide. Report No. FHWA-HRT-04-091. FHWA, U.S. Department of Transportation, 2004. 7. Messer, C. Extension and Application of Prosser-Dunne Model to Traffic Operation Analysis of Oversaturated, Closely Spaced Signal- ized Intersections. Transportation Research Record 1646, TRB, National Research Council, Washington, D.C., 1998, pp. 106–114. 8. Messer, C. Simulation Studies of Traffic Operations at Saturated, Closely Spaced Signalized Intersections. Transportation Research Record 1646, TRB, National Research Council, Washington, D.C., 1998, pp. 115–123. 9. Rouphail, N., and R. Akcelik. A Preliminary Model of Queue Inter- action at Signalized Paired Intersections. Proc., 16th ARRB Confer- ence, Australian Road Research Board, Vermont South, Victoria, Australia, Vol. 5, 1992, pp. 325–345. 10. Quinn, H. A Review of Queue Management Strategies. ITS University of Leeds. Deliverable No. 1 for Workpackage No. WP110. July 1992. 11. Prosser, N., and M. Dunne. A Procedure for Estimating Movement Capacities at Signalized Paired Intersections. Presented at Second International Symposium on Highway Capacity, Sydney, Australia, 1994. 12. Papacostas, C. S., and P. D. Prevedouros. Transportation Engineer- ing and Planning. Prentice Hall, Inc., Englewood Cliffs, N.J., 1993, pp. 307–310. 13. Hecker, J. Reducing Congestion: Congestion Pricing Has Promise for Improving Use of Transportation Infrastructure. GAO-03- 735T. U.S. General Accounting Office, 2003. 14. Transportation Research Board. Access Management Manual. Transportation Research Board of the National Academies, Wash- ington, D.C., 2003. 15. Douglas, J. G. NCHRP Synthesis of Highway Practice 314: Strategies for Managing Increasing Truck Traffic. Transportation Research Board of the National Academies, Washington, D.C., 2003.