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.
3This chapter addresses the causes of nonrecurrent congestion, illustrates how nonre- current congestion is related to delay and reliability, and describes the costs associated with both delay and reliability. NONRECURRENT CONGESTION Traffic operational delay to motorists results from both recurrent and nonrecurrent congestion. The primary difference between recurrent and nonrecurrent congestion is one of predictability. Recurrent congestion is predictable and typically occurs during the morning and evening peak hours. It is primarily caused by inadequate base capacity of the roadway during these time periods when demand is at its highest. Drivers who embark on trips during peak hour periods expect slower travel times and plan their departure and arrival times accordingly. Nonrecurrent congestion, however, cannot be planned for because it is unpredict- able. It results from random or unplanned events, varies from day to day and from one incident to the next, and creates unreliable travel times. Nonrecurrent congestion often causes the most frustration for drivers because the longer-than-expected trip time can lead to late deliveries, missed flights, delayed meeting times, and other undesirable consequences. Although recurrent and nonrecurrent congestion each contribute to delay and play a unique role in travel time reliability, the focus here is on treatments that address nonrecurrent congestion. 1 INTRODUCTION TO NONRECURRENT CONGESTION AND TRAVEL TIME RELIABILITY
4DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION There are six main causes of nonrecurrent congestion: ⢠Traffic incidents ⢠Severe weather ⢠Special events ⢠Work zones ⢠Demand fluctuations ⢠Traffic control devices Each of these causes has specific characteristics that should be considered when attempting to quantify the nonrecurrent congestion they cause and the extent to which the design treatments presented in this guide may reduce it. Some of the important considerations associated with each of the six causes of nonrecurrent congestion are presented below. Traffic Incidents When most people hear the term traffic incidents, they think about crashes. Crashes are indeed traffic incidents, but they make up only a portion of all incidents. Incidents also include such events as stalled vehicles in a travel lane or on the shoulder, debris in the roadway, and the ârubberneckingâ that occurs when drivers slow down to observe a crash or anything else that may catch their eye. Incidents cause congestion in two primary ways. First, they tend to block lanes, reducing the capacity of the roadway for everyone trying to drive by. Even incidents confined to the shoulder slightly reduce the capacity of the highway, as drivers tend to slow down when passing a vehicle on the shoulder. Second, incidents tend to cause other drivers to slow down and look, which decreases the operating speed of the road- way and also reduces capacity. The number of lanes blocked, the amount of time the lanes are blocked, and the rubbernecking effect tend to increase when emergency responders arrive on the scene, especially for more severe incidents. The design treatments most suited to reduce delay caused by incidents are those designed to reduce the frequency of incidents, minimize the number of lanes they block, reduce the time the lanes are blocked, and shield the incident from the view of passing drivers. Severe Weather Driver behavior changes when weather conditions change. Rain and snow can cause slick pavement that may increase the frequency of crashes and slow traffic. Heavy rain or snow, blowing snow or sand, and fog can reduce visibility, which also increases the frequency of crashes and slows traffic. Heavy snow may cover the roadway, hiding lane delineations and significantly reducing the capacity of the roadway. Although no design treatment is available to reduce the frequency of these weather incidents, some of the design treatments presented in this guide can reduce the impact of severe weather. Some treatments help prevent snow or sand from blowing across the road- way, improving visibility and keeping lanes clear. Other treatments treat the roadway
5DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION surface before a storm to limit accumulation of snow and ice. Some design treatments increase the speed and efficiency of lane-clearing practices to get roadways operating at full capacity as soon as possible during and after a weather event. Special Events Special events that cause nonrecurrent congestion often happen near arenas, stadiums, conference centers, and other large gathering spaces that host sporting events, con- certs, and large meetings. Special events tend to create large demand surges during short periods of time, usually just before the event begins and just after it ends. Treat- ments to help alleviate such congestion include temporarily increasing capacity in the major direction of travel, diverting demand to alternative routes, and reducing conges- tion related to weaving (frequent lane changes near merge and diverge areas) by limit- ing freeway entrance and exit options. Work Zones Work zones are sections of the roadway or roadside on which construction, main- tenance, or utility work activities are taking place. The degree to which work zones affect the reliability of the roadway is influenced by the length of time the work zone is in place, the number of lanes closed by the work zone, the specific hours of the day during which work is being performed, and the degree to which drivers are aware of the work zone and can plan for associated delays or choose an alternative route. Short-term work zones can have a major impact on nonrecurrent congestion because they are often unexpected, and drivers are not able to plan their trips to account for the delay they cause. The first few days of a longer-term work zone, especially an emergency work zone or a work zone that was not well advertised to the public, may have a similar impact on traffic as a short-term work zone. However, the longer the work zone is present, the more drivers become aware of it and become familiar with shifted traffic patterns, alternative routes, and typical delays. As some drivers choose alternative routes, demand through the work zone is reduced, along with the delay caused by the work zone. In a long-term work zone, traffic patterns eventually reach a new equilibrium of lower demand and longer travel times than the prework zone condition, and once this occurs, drivers can again accurately estimate trip length. Although delay is still incurred, it becomes more like the delay from recurrent conges- tion, which drivers expect and plan for. Identifying the point at which congestion from work zones shifts from behaving like nonrecurrent congestion and having a negative impact on reliability to behaving more like recurrent congestion and regaining some of the lost reliability is not an exact science. It will depend greatly on the availability of alternative routes, driver familiarity with alternative routes, public awareness efforts made before the work zone was established, and other factors. The impact of other causes of nonrecurrent congestion can be amplified in a work zone. For example, work zones often limit access to shoulders or pulloff areas, which can increase the lane-blocking time of crashes and other incidents that occur within the work zones. Thus, even in long-term work zones in which a new equilibrium of demand and travel time has been reached, delay and reliability may be affected by
6DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION incidents, special events, and weather more than they would have been before the pres- ence of the work zone. Design treatments aimed at reducing the impact of work zones typically accom- plish one of three things: ⢠Minimize the capacity lost due to the work zone. This can be done by providing shoulder areas or pulloff locations, âborrowingâ capacity from the other direction of travel during work zone hours, and employing practices that allow work to be accomplished on one lane at a time. ⢠Reduce the time duration of work-zone lane closures. Depending on the work zone, this could be done by limiting lane closures to off-peak periods or by accel- erating work schedules to finish the work as quickly as possible. ⢠Reduce the number of incidents in the work zone. Safe practices for managing entrances and exits of work equipment from the work zone may limit work zoneâ related crashes. In addition, providing locations for vehicles to move in the case of a crash or breakdown may reduce secondary crashes and rubbernecking in the work zone. Many of the treatments used for minimizing the impact of work zones on con- gestion are operational in nature but may require geometric design considerations to implement. For example, the acceleration of a work zone schedule may require that alternative routes are improved or that separate express lanes are built to handle traffic while the existing facility is completely closed. Demand Fluctuations Demand fluctuation refers to the day-to-day variability in traffic demand that leads to higher traffic volumes on some days than on others. It also includes the seasonal and weekend spikes in demand that are experienced near vacation destinations. Another type of demand fluctuation that could be considered is that experienced during major evacuations. Treatments that can help address fluctuating demand include those that can temporarily increase capacity in the direction of the demand surge and those that divert demand to alternative routes. Traffic Control Devices Traffic control devices include traffic signals, ramp meters, speed limits, pavement markings, lane-use signs, and others. An example of nonrecurrent congestion due to a traffic control device that many drivers have experienced occurs when a traffic signal malfunctions and goes to flashing operation or falls âout of planâ during a high- demand period. Traffic may back up several blocks around a signal that is not functioning as it should. However, this guide focuses on treatments that address con- gestion on freeways where traffic signals are not used. Traffic control devices can cause nonrecurrent congestion on freeways when traffic management devices such as ramp meters, lane-use signs, or variable speed limit signs fail or malfunction. These treatments typically fall within the category of intelligent transportation systems, and are therefore not covered in this guide.
7DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION TRAVEL TIME RELIABILITY The more often nonrecurrent congestion occurs, the less reliable the roadway is, and the more often drivers find that the time they have allocated for their trip is incorrect. For roadways that experience frequent nonrecurrent congestion, drivers often have to allocate more time to their trip than is normally required to ensure that they arrive at their destinations on time most of the time. Each driver may have a different tolerance limit for the number of times and amount of time it is acceptable to be late, but for all drivers making this calculation, the result is that they arrive at their destinations early for all of the times when nonrecurrent congestion does not occur. This buffer time that is built into driversâ estimates of how long it will take them to reach their destinations is wasted time in the sense that it cannot be allocated to the activities the drivers would prefer to be doing. However, if drivers do not build in this buffer, they will sometimes arrive late at their destinations, which can have consequences ranging from annoyance to significant financial loss, perhaps due to a missed flight or the disruption of a manu- facturing process when a freight delivery is late. The reliability of a roadway is related to the variance of the travel times experi- enced by drivers who travel along it. Reliable roadways have travel times with low variability; that is, most travel times are very similar to one another. For example, a roadway would be considered highly reliable if travel times ranged from 30 to 33 min 95% of the time. In contrast, a roadway with travel times ranging from 30 to 45 min 90% of the time, with even greater travel times 10% of the time, would not be con- sidered as reliable. Delay from nonrecurrent congestion contributes to a decline in reliability because it increases the variability of trip time, but delay from recurrent con- gestion does not affect reliability as much because the delay is fairly constant and pre- dictable. If a 5-mi trip on a freeway with a speed limit of 60 mph usually takes about 10 min during the evening peak, drivers are incurring delay due to recurrent conges- tion. However, if that travel time rarely varies from its 10-min average, the roadway is still reliable, because that trip time can be accurately planned for. If incidents are fairly common along the roadway that increase the delay on random days, causing the trip to take 15 or 20 min once or twice a week, the roadway is less reliable. Reliability is an important component of roadway performance and, perhaps more importantly, of motorist perception of roadway performance. Reliability has only recently been gaining recognition as an important performance measure for road- ways, but increasingly agencies are using reliability measures to assess their own per- formance and to communicate it to the public. Researchers have noted several reasons why measuring and managing reliability is important. These include the following: ⢠Motorists have less tolerance for unexpected delay than for expected delay. Motor ists are frustrated by not knowing and being able to plan for unexpected de- lay. Motorists want to know what to expect, and knowing the extent and duration of congestion not only gives the motorist better options, it removes a significant stress pointâthe unknown. ⢠There are costs associated with planning for unreliable travel. Travelers who use particular highways regularly (e.g., commuters, delivery services) develop an
8DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION intuitive understanding of how events cause unexpected delay. As a result, they will pad their budgeted travel time to account for the possibility of unexpected delay so that they will arrive at their destinations on time (even if this means arriv ing early on some days). This padding or buffering has not been included in economic analyses of transportation, yet it is clear that it carries a cost. For ex- ample, the additional time budgeted for travel (i.e., buffer time) is time that could have been spent doing more desirable activities. Also, the cost of being late (due to extreme events that cause the actual time to exceed the buffer), is probably valued highly by travelers. ⢠Reliability is a valued service in other utilities and industries. Ensuring that ser- vices delivered to customers are dependable and consistent is a hallmark of many nontransportation industries. Transportation services provided by public agencies should be no different. ⢠Reliability can be improved by implementing strategies to address nonrecur- rent congestion. By addressing the causes of delay, reliability can be improved. Transportation operations strategies such as incident, weather, and work zone management have proven to be cost-effective strategies for addressing congestion. Therefore, reliability measurement provides a glimpse into how well transporta- tion operations strategies are working. COSTS OF DELAY AND RELIABILITY Many drivers will choose a path with reliable congestion over a path with highly unre- liable travel times, even if the unreliable path will occasionally get them to their desti- nation faster than the reliable one. This indicates that reliability has a value to drivers separate from the value of time and that reliability should be a factor in benefitâcost analyses of projects that may affect it. It is well established that drivers incur a cost when they experience delay. The time lost due to delay on the highway can be assigned a monetary value, which is often mea- sured in terms of lost productivity, represented by average hourly wages. These values can be calculated for delay caused by both recurrent and nonrecurrent congestion. The idea that the reliability of a roadway can also be assigned a cost is a newer concept, but one that is gaining acceptance among transportation planners. At present, there is no commonly accepted way to value reliability monetarily, but methods continue to be suggested, tested, and applied in current research. The most frequently used approach is to define a reliability ratio, which relates the value of reliability to the value of time. Reliability ratios that have been used in recent literature range from approximately 0.7 (meaning that reliability costs 70% as much as time) to 1.3 (meaning reliability costs 1.3 times the value of time). Different reliability ratios may be used for different trip types (such as commute, noncommute, and freight). The benefits provided by the treatments included in this guide will likely vary widely from one application to another. For some, design options or constraints may affect the benefits available from the treatment, but for others, local policies and
9DESIGN GUIDE FOR ADDRESSING NONRECURRENT CONGESTION practices may affect the frequency of use and effectiveness of the treatment. As an example, crash investigation sites are affected in both ways. The design, location, and prevalence of such sites, which can range from extra-wide shoulders to paved and lit locations in the median of the freeway, affect how likely motorists are to use the sites. In addition, the policy of the emergency responders, as well as public awareness cam- paigns and signing, also affect how frequently the sites are used, and therefore, their effectiveness and benefits. The factors that are likely to affect the effectiveness of the treatments are included in the discussion for each treatment later in this guide. Road- way planners and designers should consider these factors when estimating the benefits they will achieve by implementing any given treatment. REDUCING NONRECURRENT CONGESTION AND IMPROVING RELIABILITY When nonrecurrent congestion is minimized, travel times become more predictable and reliability is improved. The remainder of this guide is dedicated to presenting several highway design treatments that use various mechanisms to mitigate conges- tion and delay due to incidents, weather events, special events, work zones, demand fluctuations, and traffic control devices. The following chapters provide detailed in- formation about how to choose appropriate treatments and how to maximize their effectiveness.
