System Performance Defined

As used here, system performance refers to how efficiently and reliably people and freight carriers can travel to destinations on the transportation network. This chapter focuses on measures that can be used to determine whether certain aspects of system performance, such as congestion and accessibility, are improving or declining over time. Other aspects of system performance, such as safety, energy usage, and environmental impacts, are discussed separately in other chapters.

System performance measures often are viewed from the perspectives of both the user and the operator. Users are interested in characteristics, such as travel cost, travel time, and the reliability of successfully completing a trip within a certain time, each of which directly affects their ability to accomplish a trip purpose. Owners and operators are concerned with the level of service provided to users and the ability to respond to service disruptions so as to promote reliable mobility and accessibility.

The U.S. Congress has recognized the importance of system performance by calling for performance-based decision making in MAP-21 (Moving Ahead for Progress in the 21st Century Act), signed into law by President Obama in July 2012 [Pub. L. 112–141].

As part of MAP-21, the U.S. Department of Transportation (USDOT) is required to establish performance measures and standards for several program/policy areas, including asset conditions on National Highway System (NHS) roads, safety, mobile source emissions, performance of interstate and noninterstate NHS roads, traffic congestion, and freight movement on the interstate system. These latter three categories—interstate/noninterstate NHS road performance, traffic congestion, and freight movement—relate most directly to the performance of the Nation’s transportation system and are thus the focus of this chapter. MAP-21 also requires statewide and metropolitan transportation planning agencies to establish and use performance-based approaches for transportation decision making that support national goals (see box 5-A).

System Accessibility

System accessibility is defined as the ability of travelers and freight shippers and carriers to reach key destinations, such as hospitals, job sites, schools, factories, airports, ports, and community centers. The Nation’s transportation infrastructure is comprised of over 4 million miles of roads, about 19,400 public and private use airports, 140,000 miles of freight and passenger railroads, and 25,000 miles of navigable waterways as shown in box 1-A. In 2014 the U.S. transportation system served 316.1 million residents as well as millions of businesses and visitors [USDOC CENSUS 2014a]. In today’s world, one can accomplish many objectives without ever traveling, such as electronic banking, shopping, and communications. This substitution effect for trip-making has in some cases reduced the number of trips made, but it might increase the number of trips in other categories, such as the number of truck deliveries in response to Internet shopping.

In evaluating system performance, it is important to know how accessibility has changed over time. The measure most often used is the number of destinations reachable within a given travel time, in particular transportation system accessibility to jobs. The Center for Transportation Studies, at the University of Minnesota, has developed a method for comparing morning peak-period accessibility to jobs by automobile across 51 U.S. metropolitan areas for 1990, 2000, and 2010 [UMN CTS 2013]. Figure 5-1 shows how accessibility to jobs has changed from 1990 to 2010. In 1990, for example, 2 million jobs across 51 metropolitan areas were accessible in an average travel time of 43 minutes by automobile. A decade later, in 2000, the average travel time had increased to 52 minutes. But by 2010 that average travel time had dropped to 46 minutes as travel speeds increased (to about where they were in 1990) [UMN CTS 2013]. The crossing of the 2000 and 2010 lines in figure 5-1 most likely reflects the impact of the December 2007 through June 2009 recession and the subsequent slow recovery, that is, not as many jobs were available for access.

The University of Minnesota study concluded that increases in absolute accessibility were found in large (Dallas, San Francisco, and Los Angeles) and smaller (Jacksonville, Las Vegas, and Phoenix) fast-growing metropolitan areas. Decreases in accessibility for the period of 1990 to 2010 tended to be found in older northeastern metropolitan areas that had minimal or negative job growth and faster-growing cities with major congestion problems. Changes in accessibility were attributed to changes in network speeds, network design, and employment density. It is important to emphasize that this accessibility measure focused only on trips by automobile; a complete accessibility measure would include accessibility from transit and other modes.

Congestion

The ability of transportation system users to reach a destination in a cost-effective, safe, and reliable manner is an important aspect of the Nation’s transportation system. The characteristics of making such trips, including travel time, costs, and access to facilities/ services, are used to indicate the level of mobility afforded to users. Box 5-B describes how system performance measures, such as travel time and congestion, are viewed from two different perspectives—the user’s versus the operator’s.

Road congestion in urban areas is one of the major causes for travel time delay. The Texas Transportation Institute has monitored congestion levels on the U.S. road network for decades and has reported in a biannual Urban Mobility Report on the number of hours of congestion experienced by network users and the associated economic costs [TAMU TTI 2013].

Table 5-1 shows the estimates for annual hours of delay, the number of gallons of wasted fuel due to delay, the dollar value of delay and wasted fuel, and a measure called the Travel Time Index (TTI).¹ For example, a TTI value of 1.18 indicates that a 30 minute trip with congestion will take 18.0 percent longer or just over 35 minutes (1.18 × 30) in the peak period.

Road congestion, in terms of amount and cost, has increased since 1990, although the economic recession that began at the end of 2007 and ran through the middle of 2009 has had a dampening effect on what had been a steady increase. Congestion in the Nation’s urban areas in 2011 had an economic cost of $121.2 billion compared to $55.2 billion in 1990 (2011 dollars). The average yearly delay per commuter rose from 32 hours in 1990 to 38 hours in 2011, a 19 percent increase, and the total national hours of delay in 2011 reached 5.5 billion hours—more than twice the 1990 total. The high points of congestion, however, were in 2005, 2006, and into 2007, just before the recession began. The effects of congestion on truck movements accounted for $27 billion (22 percent) of the congestion cost [TAMU TTI 2013]. In addition, the average commuter:

• wasted 19 gallons of fuel per week in 2011 (a week’s worth of fuel for the average U.S. driver), up from 8 gallons in 1982;
• experienced an average of 52 hours of delay in 2011 in areas with over three million population; and
• planned for approximately 3 times as much travel time as in noncongested conditions to arrive at their destination ontime 9 times out of 10 [TAMU TTI 2013].

In all urban areas, the worst congestion levels (defined as “extreme,” “severe,” or “heavy”) affected only one in nine trips in 1982, whereas this proportion increased to almost one in three trips in 2011. In addition, the most congested sections of road (labeled extreme and severe) handled only 21.0 percent of all urban road travel, but accounted for 78.0 percent of peak period delays as shown in figure 5-2. It is important to note that congestion levels have increased over the past 30 years in all urban areas, from the largest to the smallest. Congestion is worse in the afternoon, but it can occur at any time throughout the day (figure 5-3).

Congestion and delay are not limited to roadways. The average length of flight delays is over 50 minutes (table 5-2). Flight delays are caused by a variety of reasons, ranging from extreme weather to disruptions in airline carrier operations (figure 5-4). The combined effects of nonextreme weather conditions, airport operations, heavy traffic volume, and air traffic control contributed to 22.5 percent of delays in 2012, an 11 percentage point improvement from 2005. Flight delays can ripple through the U.S. aviation system as late arriving flights, for whatever reason, delay subsequent flights—the cause of 42.1 percent of delays for the next scheduled flights in 2013.

Congestion is especially a problem for time-sensitive freight shipments. Various performance indicators are used to monitor time-related system performance. The USDOT’s Federal Highway Administration (FHWA), in cooperation with the American Transportation Research Institute (ATRI), is working to quantify the impact of traffic congestion on truck-based freight at 250 specific locations. Similar to the TTI, the primary measure is the ratio of noncongested speed to congested speed at key freight locations (often interstate-to-interstate interchanges). For example, a 22.6 mph peak average speed and a 43.4 mph nonpeak average speed in Austin, TX, yields a ratio of 2.10. In 2012 some of the most congested truck bottlenecks on freight-heavy highways could be found in Austin, TX (2.10); Chicago, IL (1.87); Houston, TX (1.51); and Atlanta, GA (1.49) [USDOT FHWA and BTS 2013].

The inland water network is also a key component of the Nation’s freight transportation system. The U.S. Army Corps of Engineers (Corps) is responsible for 239 lock chambers on the Nation’s inland water system and monitors the movements of barges and other commercial vessels. In 2012 barge tows experienced on average a 2-hour delay navigating a lock, the largest delay since 2000 [USACE 2012]. The average age of locks under jurisdiction of the Corps is 50 years, and it is expected that delays will likely increase given the needed rehabilitation and reconstruction of key locks.

System Reliability

Reliability is defined as the level to which one can make trips with some certainty that the actual trip will occur within an expected range of travel times. More reliability means less uncertainty associated with trips, such as crashes, vehicle breakdowns, and similar incidents; work zones; unannounced road work; weather; and special events that can often lead to widely varying travel times from one day to the next for the same trip.

The Planning Time Index (PTI)² is used to estimate the extra time that one should plan for a trip. For example, a PTI of 1.5 means that a traveler wanting to arrive on time 19 out of 20 times should allow 50 percent more time. This means 30 extra minutes should be budgeted for a trip that in free flow conditions would typically take 60 minutes to arrive on time in 19 out of 20 times. The extra time allowed, in this example 30 minutes, is called the buffer index, which is often used to assess system reliability. Based on PTI data collected from 18 cities between 2007 and 2012, travelers would have to plan a minimum of about 40 percent more travel time to arrive “ontime” for 19 out of 20 trips. Significant progress was made to reduce roadway congestion from 2007 to 2012 (figure 5-5) [USDOT FHWA OPS 2013]. Figure 5-5 also shows the impact of weather on travel as the PTI was generally higher in winter than in summer months.

For nonhighway modes, different measures can be used to assess system reliability. For passenger transportation, for example, ontime performance is often an indicator of service reliability. Amtrak experienced a significant improvement in ontime performance, with a record 83.0 percent ontime performance in 2012 [USDOT BTS NTS 2013], up from 68.6 percent in 2007 (table 5-3). Greater improvement in ontime performance is seen for trips over 400 miles in length, where ontime performance jumped from 39.5 percent in 2007 to 70.7 percent in 2012. The vast majority of passenger train services outside the Northeast Corridor are provided over tracks owned by and shared with the Class I freight railroads. As a result, Amtrak’s ontime performance is largely dependent on the condition and performance of the host railroads, with the important exception of Amtrak-owned tracks in the Northeast Corridor.

U.S. airlines reported that nearly 20 percent of all domestic scheduled flights, or more than one million flights, arrived at the gate more than 15 minutes late in 2013. The average length of delay for late arriving flights was almost an hour. More than 10 percent, or nearly 126 thousand flights, arrived at the gate more than 2 hours late (table 5-2). Between 2005 and 2012, late departures decreased from 17.9 (for all carriers and all airports) to 16.3 percent, while late arrivals declined from 20.5 to 16.6 percent [USDOT BTS NTS 2013].

For the U.S. Army Corps of Engineers inland waterway locks, system reliability can be measured as the percent of time a lock is unavailable for use (defined as the cumulative periods over a year during which a lock facility was unable to pass traffic). Locks could be unavailable for a number of reasons, ranging from scheduled maintenance, unexpected stoppages due to operational issues, and weather conditions such as flooding and ice. For example, high water levels and flows shut down 22 locks and stopped cargo movements along the Upper Mississippi River and its confluences in late April 2013 [USACE 2013]. As shown in figure 5-6, the total number of hours of unavailability in 2012 was more than 150 thousand, almost twice the level in 2000. Lock unavailability due to scheduled operations, such as maintenance, showed a marked increase, accounting for 84.7 percent of the total unavailable hours in 2012. As noted earlier, the increase is most likely due to the aging of the locks in the inland water system.

System Resiliency

Many parts of the Nation’s transportation system are vulnerable to both natural and man- made disruptions. Because of this vulnerability, transportation firms and agencies have become interested in providing a system that is resilient to disruptive impacts. A resilient transportation system has design-level robustness so that it can withstand severe blows, respond appropriately to threats, and mitigate the consequences of threats through response and recovery operations [USDOT VOLPE 2013]. A resilient transportation system is one that can “take a punch” and recover in a timely way to provide the mobility and accessibility that are critical to the economy and to the quality of life of the Nation’s citizens.

The United States has experienced extreme weather events throughout its history. However, with the heavy concentration of the Nation’s population in urban areas (many along the coasts) and with a strong reliance on the efficient movement of people and goods, recent weather events have resulted in extensive economic and community costs. For example, the U.S. Department of Commerce (USDOC), National Oceanic and Atmospheric Administration (NOAA) estimated that the United States has experienced 151 weather/ climate disasters since 1980, including such events as hurricanes, tornadoes, floods, and droughts/wildfires. The overall damage from each of these events exceeded $1 billion, resulting in more than a $1 trillion cumulative cost to the Nation [USDOC NOAA 2013a]. The year 2012 was the second most costly since 1980 with over $110 billion in damages and 377 deaths due to extreme weather events (dollar wise, the most costly year was 2005 with over $160 billion in damages) [USDOC NOAA 2013b]. Part of the physical recovery costs and overall economic impact were due to the damage and disruption to the transportation system.

Hurricane Sandy and Tropical Storm Irene are two recent examples of extreme weather events that disrupted the transportation system. Hurricane Sandy caused extensive damage in October 2012 along the New Jersey, New York, and Connecticut coasts and record flooding in lower Manhattan. Roads and bridges were damaged throughout the region, and road and rail tunnels were flooded. The region’s major airports were closed, and transit service was not restored in many areas until several months after the storm [KAUFMAN, QING, LEVENSON and HANSON 2012].

Tropical Storm Irene, the remnants of what had been Hurricane Irene that hit the U.S. East Coast in 2011, inundated Vermont with high levels of rainfall. Over 225 communities were affected by the storm. Massive flooding closed 146 segments of the state road system with over 200 bridges damaged (34 closed completely), resulting in a repair cost of between $175 million to $200 million. An additional 2,260 municipal road segments and 963 culverts were damaged, with 175 road segments and 90 bridges closed. Moreover, 200 miles of state-owned rail track required repair [USDOC EDA 2012].

Although the New Jersey/New York/ Connecticut and Vermont regions suffered huge losses during their respective storms, one of the key lessons from each event was the importance of transportation system resilience. Major transportation facilities—roads, bridges, transit systems, ports, and airports—were in operation within weeks of the hurricane. In most cases, advanced preparations by state and local government agencies (e.g., moving transit vehicles out of vulnerable areas and establishing emergency management centers) can mitigate disruption to transportation systems [MTA 2012]. The existence of redundant paths in the New Jersey/New York/ Connecticut and Vermont transportation network provided travel options for both person and freight trips seeking to avoid travel blockages. In both cases, the transportation agencies were able to put the transportation system back into operation in a very short period of time, thus minimizing the economic impact to state and regional economies.

There are economic and other costs associated with such major disruptions, including those resulting from extreme weather events, infrastructure repair, and loss in productivity. For example, the economic impact to New Jersey and New York resulting from Hurricane Sandy was estimated at $63 billion, although some studies have suggested that the impact was less given the economic rebound associated with the recovery from the hurricane [RUTGERS UNIVERSITY 2013]. This cost included the estimated expenditures to replace the roads, bridges, and transit facilities damaged by the storm.

A Washington State DOT study of the economic impact of massive floods that closed I-5 and I-90 during the winter of 2007–2008 estimated a combined cost of almost $75 million, of which some $47 million was associated with the I-5 disruption [PSRC 2008]. The analysis was based on revenue losses incurred by firms that could not deliver products to or receive orders from customers and on additional business costs incurred by both the trucking industry and freight dependent sectors because of delays, detours, and the use of alternative modes of delivery. The additional cost of taking the detours was estimated to be between $500 and $850 per truckload. The Washington State DOT estimated the loss of more than $3.8 million in tax revenues and 460 jobs. In addition to the above business losses, estimated highway damage from the winter storm totaled $18 million for state routes and another $39 million for city and county roads [WSDOT 2008]. The U.S. Department of Transportation, in 2014, projected the potential impacts of changing climatic conditions and associated infrastructure costs on Mobile, AL. Figure 5-7 shows the likely inundation levels in the Mobile region and the critical transportation facilities that would be affected with a Hurricane Katrina-like storm on top of a 0.8 yard (0.75 meter) sea level rise. The areas in various shades of red indicate the water-level elevation over current dry land and the location of critical transportation facilities. As can be seen, a significant amount of the Mobile region would be inundated, along with several key transportation facilities. There is a significant concentration of fuel ports and terminals such as the Louisiana Offshore Oil Port and 33 refineries along gulf coast. Many of these ports and terminals are connected to an aging network of oil and gas pipelines, half of which were built 50 to 60 years ago, that are susceptible to floods and wind damage [USDOT FHWA 2014].

Security Concerns

The Transportation Security Administration (TSA) of the U.S. Department of Homeland Security screens people as they pass through security checkpoints at 450 airports with Federal screening and other passenger facilities. The TSA confiscated approximately 50 million prohibited items, such as sharp objects, firearms, tools, explosive and flammable material, volatile chemicals, and other dangerous items, over the past decade. In 2011 alone, the TSA prevented more than 1,200 guns from being brought onto passenger aircraft [USDHS TSA 2012].

International piracy incidents and armed robberies at sea are another security concern affecting U.S. citizens traveling overseas, especially in the waters surrounding the horn of Africa. This area has been monitored closely, especially after the hijacking of the U.S.-flagged Maersk Alabama on April 8, 2009. In 2013 there were 9 vessels fired upon/ attempted boarding in East African waters compared to 7 vessels hijacked, 1 boarding, and 24 vessels fired upon/attempted boarding in 2012 [USN ONI 2014].

Cost Associated With Poor System Performance

The following discussion focuses on the economic costs associated with poor transportation system performance, costs associated with system disruptions, and expected benefits from strategies that will improve system performance.

The Urban Mobility Report includes an estimate of the cost to system users of about $121 billion in delay and fuel wasted in congestion costs in 2011. The report also estimated the beneficial effects of public transportation and roadway operational improvements to these costs. For public transportation, the analysis examined what would happen if transit services were eliminated in the 498 urban areas that were part of the study. The additional system cost (or the cost foregone given transit service) is thus considered the benefit of transit investment. For 2011 the benefit includes 865 million hours of delay eliminated and 450 million gallons of fuel saved, resulting in an estimated $20.8 billion in cost savings [TAMU TTI 2013]. For road operational improvements, the report estimated 364 million hours of delay eliminated and 194 million gallons of fuel saved, resulting in an estimated $8.5 billion in cost savings [TAMU TTI 2013].

With respect to businesses, three critical aspects of operations can be affected directly by congestion:

1. direct travel (user) cost, including vehicle operating costs and value of time for drivers and passengers, for all business- related travel;
2. logistics and scheduling costs, including costs of stocking, perishability, and just-in- time processing; and
3. market accessibility and scale, including loss of market-scale economies and reduced access to specialized labor and materials because of congestion.

With a new emphasis on performance-based decision making from MAP-21, it is likely that state transportation planning agencies throughout the Nation will be collecting more data on system performance. This data, and the information it produces, could be useful to decision makers in identifying targeted opportunities for improving transportation system performance, with its attendant economic and quality of life benefits.

References

Kaufman, S., C. Qing, N. Levenson and M. Hanson. Rudin Center for Transportation, New York University Wagner Graduate School of Public Service. 2012. Transportation During and After Hurricane Sandy (November 15, 2012). Available at http://wagner.nyu.edu/ as of December 2013.

Metropolitan Transportation Authority (MTA). MTA Prepares for Hurricane Sandy (October 26. 2012) . Available at http://www.mta.info as of January 2014.

Puget Sound Regional Council (PSRC). Freight Transportation Economic Impact Assessment of the I-5 and I-90 Closures in 2007-2008 (July 11, 2008). Freight Mobility Roundtable. Presentation. Ivanov, B., Hammond, P. and Reinmuth, S. Seattle, WA.

Rutgers University. The Economic and Fiscal Impacts of Hurricane Sandy in New Jersey: A Macroeconomic Analysis, Issue Paper No. 34. (January 2013). Available at http://policy. rutgers.edu/ as of December 2013.

Texas A&M University (TAMU), Texas Transportation Institute (TTI). 2012 Urban Mobility Report (2013). Available at http://tti. tamu.edu/ as of December 2013.

University of Minnesota (UMN), Center for Transportation Studies (CTS). Access Across America (2013). Report CTS 13-20. Available at http://www.cts.umn.edu/ as of December 2013.

U.S. Army Corps of Engineers (USACE), Navigation Data Center:

—2013. Lock closures (as of April 23, 2013). Personal Communications.

—2012. Locks by Waterway, Lock Usage, CY 1993 – 2012. Available at http://www. navigationdatacenter.us as of December 2013.

U.S. Department of Commerce (USDOC). Census Bureau (CENSUS).

—2014a. American Fact Finder. Available at http://factfinder2.census.gov/ as of July 2014.

—2014b. Mean Travel Time to Work. Available at http://www.census.gov/ as of January 2014.

—2013. Commuting (Journey to Work). Available at http://www.census.gov/ as of October 2013.

U.S. Department of Commerce (USDOC). Economic Development Administration (EDA). Economic Impact Assessment, Economic Recovery Support Function. Vermont DR- 4022 (April 2012). Available at http://vtstrong. vermont.gov/ as of December 2013.

U.S. Department of Commerce (USDOC). National Oceanic and Atmospheric Administration (NOAA). National Climatic Data Center (NCDC).

—2013a. Billion-Dollar Weather/Climate Disasters. Available at http://www.ncdc.noaa. gov/billions/ as of January 2014.

—2013b. National Oceanic and Atmospheric Administration (NOAA), National Climatic Data Center (NCDC). NCDC Releases 2012 Billion-Dollar Weather and Climate Disasters Information. Available at http://www.ncdc.noaa. gov/billions/ as of January 2014.

U.S. Department of Transportation (USDOT). Bureau of Transportation Statistics (BTS). National Transportation Statistics (NTS). Available at http://www.transtats.bts.gov/ as of December 2013.

U.S. Department of Transportation (USDOT). Federal Highway Administration (FHWA) and Bureau of Transportation Statistics (BTS). Freight Facts and Figures 2013. Available at http://www.ops.fhwa.dot.gov/ as of June 2014.

U.S. Department of Transportation (USDOT). Federal Highway Administration (FHWA), Impacts of Climate Change and Varability on Transportation Systems and Infrastructure: The Gulf Coast Study, Phase 2, Task 3.1 (June 2014), Report FHWA-HEP-14-033. Available at http://www.fhwa.dot.gov/ as of October 2014.

U.S. Department of Transportation (USDOT), Federal Highway Administration (FHWA), Office of Operations (OPS). Urban Congestion Report (UCR): January 2013 through March 2013. Available at http://www.ops.fhwa.dot.gov as of December 2013.

U.S. Department of Transportation (USDOT), Volpe National Transportation Center (VOLPE). Beyond Bouncing Back: Critical Transportation Infrastructure Resilience (April 2013)Available at http://www.volpe.dot.gov/ as of December 2013

U.S. Department of Homeland Security (USDHS). Transportation Security Administration (TSA). Testimony of John Pistole, Administrator of the Transportation Security Administration before the House Committee on Homeland Security, Subcommittee on Transportation Security addressing the TSA Screening Partnership Program (February 7, 2012). Available at https://www.dhs.gov as of February 2014.

U.S. Navy (USN). Office of Naval Intelligence (ONI). Horn of Africa and Gulf of Guinea: Piracy Analysis and Warning Weekly (PAWW) Report for 13 – 19 February 2014 (February 20, 2014). Available at http://www.oni.navy.mil as of February 2014.

Washington State Department of Transportation (WSDOT). 2008. Storm Related Closures of I-5 and I-90. Freight Transportation Economic Impact Assessment Report. Winter 2007-2008. Final Research Report WARD 708.1., Olympia, WA.

¹ The ratio of the travel time during the peak period to the time required to make the same trip at free-flow speeds.

²  The ratio of travel time on the worst day of the month compared to the time required to make the same trip at free-flow speeds.