Performance Measurement Framework for Highway Capacity Decision Making (2009)

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.

32C H A P T E R 4 Transportation FactorsBackground Literature Transportation agencies have been using performance mea- sures to understand the implications of their investments on the transportation system for decades, and these practices stand as a model for incorporating the other factors included in this performance measurement framework. Many perform- ance measures used to plan, operate, and monitor transporta- tion facilities today are descendents of measures conceived in the 1950s (Meyer, 2001), including: • Mobility and reliability measures: – Annual average daily traffic per lane-mile; – Average travel rate (minutes per mile); – Nonrecurring delay; – Incident-related delay; – Travel time index (median reliability measure); – Planning time index (95th percentile reliability measure); and – Percentage of vehicle miles of travel under congested conditions. • Safety measures: – Number and rate of fatalities and injuries; and – Number of crashes by type, including run-off-the- road, pedestrian,heavy-vehicle, impaired-driver, repeat- offender, uninsured-driver, and unlicensed-driver. • Infrastructure condition and deficiency measures: – Average ride quality; – Percentage of asset length or count by condition; – Remaining life; – Bridge health index; and – Bridge deck condition. These and many other traditional system-related measures are discussed at length in the literature (Cambridge Systemat- ics, Inc., 2000; Cambridge Systematics, Inc., 2007; Brydia et al., 2007; Shaw, 2003; Cambridge Systematics, Inc., 2005). Harrisonet al., (2006) compile similar performance measurements specifically for freight transportation. The remainder of this section discusses recent trends. Key Findings Performance measurement of transportation systems is increasingly operations-oriented. Performance measurement of transportation systems over the last 50 years has been crit- icized for being reflective of the values held by the engineers expanding capacity of the National Highway System (NHS) (Hendren and Myers, 2006; Meyers, 2001). But as the NHS is all but built out, the focus of engineers has shifted from construction to operations. Marginal benefits of operational improvements are typically much smaller than those of capacity improvements, thus new measures are needed to more accurately reflect travel characteristics. Recent literature strongly supports this trend (Cambridge Systematics, Inc., 2007; Brydia et al., 2007; Randall, forthcoming; Cambridge Systematics, Inc., 2005). Performance measures must be viewed from both system and user perspectives. The literature and feedback from practitioners have indicated a trend toward measures that capture how customers experience the transportation system (Hendren and Meyers, 2006). Customer-oriented perform- ance measures, including random sample surveys of travel- ers, web feedback, utilization of traveler information services, press clippings, and media editorials have been used to gauge the quality of the customer experience (Adams et al., 2005). Though these outcome measures are important, output mea- sures such as speed, travel time, delays, and number of acci- dents remain relevant. Perspectives from both the system and the user are needed (Cambridge Systematics, Inc., 2007; Adams et al., 2005; Shaw, 2003). Measures of reliability are growing in importance for both passenger and freight travel. Passengers willingly accept reasonable regular and predictable delays. Users are particularly

33sensitive, however, to unanticipated delay caused by inci- dents, construction, weather, demand fluctuations, special events, traffic control devices, or inadequate base capacity. Industry has largely turned to just-in-time production and delivery methods, and similarly it can accept anticipated delays. But variability in travel times can significantly delay shipments beyond expected delivery times (Harrison et al., 2006). Because the highway system is mature, transportation improvements result in increasingly smaller marginal improve- ments in average travel times. Thus, importance of system reliability is increasing relative to the importance of average travel time (Cambridge Systematics, Inc., 2000; Cambridge Systematics, Inc., 2007). Measures of mobility and accessibility are distinct but complementary. Mobility represents the ability to travel (e.g., travel time, travel rate); accessibility is the ability to reach desired destinations or activities. Measures of accessibility generally address issues such as: • Access of persons/households to jobs (e.g., employment opportunities within 30 minutes); • Access of employers to labor force (e.g., workers within 30 minutes); • Access of persons to other opportunities (e.g., shopping, school, daycare); and • Access of persons to alternative modes of transportation, especially transit (e.g., population within one-quarter mile of a transit stop). There are five categories of accessibility measures: 1. Isochronic (or cumulative) opportunity that measures the number of opportunities (e.g., jobs) within a given travel time; 2. Gravity-based measures that weight opportunities by time/ distance; 3. Utility-based measures that weight opportunities by their relative importance/benefit; 4. Constraints-based measures that introduce temporal feasibility; and 5. Composite accessibility measures that combine features of several of the above measures. Research efforts continue to investigate how to integrate accessibility measures into planning and monitoring efforts (Chen et al., 2007; Bhat et al., 2006; Levinson and Krizek, 2005). A new measure, “Place Rank,” is inspired from a methodol- ogy used in ranking web pages for the Google search engine. It takes advantage of origin and destination information and can be implemented without knowing point-to-point travel time (El-Geneidy and Levinson, 2006).Safety performance measurement is advanced in other developed nations. An international scan of performance- based planning found Australia, New Zealand, British Colum- bia, Canada, and Japan to have particularly advanced safety performance measurement regimes. The team contributed the successes of these countries to their ability to: • Understand underlying safety problems; • Establish institutional leadership and accountability; • Define safety performance measures and targets; • Collect and analyze data; • Benchmark results against other agencies; and • Integrate results into agency decision making (McDonald et al., 2004). In the United States, performance measurement is a key focus of the Integrated Safety Management System, a system for bolstering highway safety by integrating the disciplines of numerous public agencies (Bahar et al., 2003). Transportation Performance Factors and Measures In evaluating major capacity expansion projects, impacts on the movement of people and goods over that system are among the most common considerations. The performance mea- sures framework identifies four categories for evaluating the impact of capacity-adding projects on transportation system performance. • Mobility; • Reliability; • Accessibility; and • Safety. Mobility Mobility refers to the ability of the transportation system to facilitate efficient movement of people and goods. Mobility typically addresses recurring congestion that results when traffic volumes approach or exceed available roadway capac- ity. Mobility measures do not capture the implications of the location of the congestion compared to desired destinations, but instead simply highlight the extent of congestion in comparison with free-flow conditions. Two guiding objectives were identified for mobility: • Reduce recurring congestion – improve travel time; and • Reduce traffic volume. Table 4.1 presents seven broad performance measures to address these objectives and specific applications of each performance measure. The case study highlight illustrates

34Table 4.1. Transportation Measures – Mobility Case Study Highlight: Mn/DOT 2003 Statewide Transportation Plan Description: Minnesota’s 2003 Statewide Transportation Plan and 2005 district-level plans comprise one of the nation’s first comprehensive, performance-based state transportation planning efforts. The Statewide Plan sets a framework for long-range investment planning, with perform- ance measures and targets in 10 policy areas. The district-level plans identify investment levels needed to meet targets and detail a prioritized, fiscally constrained 20-year implementation program. The statewide and district plans serve as the critical link between Mn/DOT’s strategic goals and the capital investment program in the Statewide Transportation Improvement Program (STIP). Mn/DOT employs regular performance monitoring to evaluate investment choices and adjust the state’s investment program. Sample Measures: • Policy: Enhance Mobility in Interregional Transportation Corridors Linking Regional Trade Centers (RTC): – Travel Speed – Percent of IRC miles meeting speed targets; and – Travel-Time Reliability – Peak period travel time reliability. • Policy: Enhance Mobility Within Major RTCs: – Travel Time – Ratio of peak to off-peak travel time (Travel Rate Index); and – Travel-Time Reliability – Peak-period travel time reliability. SHRP 2 Framework Measure Specific Measure Applications Recurring Delay – Difference between the actual time required by motorist to traverse a roadway segment and the unconstrained time. Trip Travel Time – Time required for a motorist to com- plete a trip from its origin to its destination. Travel Time Index – Ratio of the actual travel time for a trip compared to the unconstrained travel time. Volume to Capacity Ratio – Actual number of vehicles using a roadway segment relative to the number of vehicles it is designed to handle over a fixed time period. Level of Service – Qualitative letter grade of highway operating conditions from A (unconstrained travel) to F (severe congestion). Vehicle Miles Traveled – Number of vehicles traveling a specified portion of the highway network over a set time multiplied by its length in miles. • Average daily traffic flow per freeway lane; • Ton-miles traveled by congestion level; • Delay per ton-mile traveled; • Lost time due to congestion (per vehicle or experienced by all vehicles); • Vehicle queuing and its relationship to overall delays; • Percentage of time average speed is below threshold value; • VMT by congestion level; and • Percentage of congested miles of state-maintained highways by area (urban, rural), functional class (interstate, priority, etc.). • VHT per capita; • VHT per employee; and • Average person hours of travel (PHT). • Mobility index [person-miles (or ton-miles) of travel/vehicle-miles of travel (PMT/VMT) times average speed]. • Percent of VMT which occurs at facilities with a V/C ratio greater than 0.71 or 0.8 (or another threshold); and • V/C by route. • Percent of highways not congested during peak hours; and • Number and percent of lane-miles congested. • Total VMT; • VMT growth relative to population, employment; • VMT per capita; • VMT per employee; • VMT within urban areas; • Average person miles of travel (PMT); • PMT per capita; • PMT per worker; and • Delay per VMT (by mode) Mode Share – Number of percent of transportation system users using non-SOV travel means (e.g., transit, bicycle, high-occupancy vehicle travel).

35how the Minnesota DOT incorporated many of these measures into their 2003 Statewide Transportation Plan. Reliability Reliability refers to the ability of users of the system to predict the amount of time it takes to make trips on the system. Reli- ability typically addresses nonrecurring congestion that results from traffic incidents (crashes, breakdowns, special events, weather, and construction). Factors that impact reliability include things such as route redundancy, incident response, and incident rates. Table 4.2 presents four general measures with examples that all support the objective of reducing non- recurring congestion. The case study highlight illustrates howthe Arizona DOT measures incident duration in their MoveAZ Transportation Plan. Several of these example measures have been adapted from the SHRP 2 project L03 Analytic Procedures for Determining the Impacts of Reliability Mitigation Strategies. Accessibility Accessibility refers to the ability of the transportation system to connect people to desired destinations through the spatial analysis of residential population, employment centers, and other service or recreation opportunities. Accessibility differs from mobility in that the measures can consider all modes,SHRP 2 Framework Measure Specific Measure Applications Reliability Index – A measure of the additional time (in minutes, percent extra time, etc.) that trips take under congestion conditions relative to uncongested or ‘normal’ conditions. On-Time Trip Reliability – Share of trips between a specific origin and destination with travel times below a designated threshold of time. Incident Duration – Average time elapsed from notifica- tion of an incident to incident clearance. Crash Analysis – Identification of high crash locations by roadway segment. • Buffer Index. The difference between the 95th percentile travel time and the average (or median) travel time, normalized by the average (or median) travel time (i.e., the percent extra time). • Travel-Time Index (TTI). The ratio of travel under congested conditions (i.e., 80th, 95th percentile of traffic flow) to another (i.e., median, mean of traffic flow). • Planning-Time Index. 95th percentile travel-time index divided by the free-flow travel-time index. • Skew Statistic. The ratio of (90th percentile travel time minus the median) divided by (the median minus the 10th percentile). • Misery index. The average of the highest five percent of travel times divided by the free-flow travel time. • Percent of trips with travel times less than 10 or 25 percent higher than the median travel time; • Percent of trips with space mean speed less than 50, 45, or 30 mph; and • Throughput Efficiency – Difference between actual average speed of vehicles traversing a roadway segment and speed at which maximum throughput occurs. • Average time elapsed from notification of an incident until all vehicles have moved to shoulder; • Average time elapsed from notification of an incident until all vehicles have been removed from scene; and • Average time elapsed from notification of an incident until all last responder has left the scene. • Location of highest crash rate (accidents per traffic volume); and • Number of locations with crash rate higher than national average (accidents per traffic volume). Case Study Highlight: Arizona DOT MoveAZ Transportation Plan Description: MoveAZ is the Arizona DOT’s current long-range transportation plan. MoveAZ was developed using a comprehensive performance- based planning effort to support a process in which needs and projects identified in the plan ultimately move to programming and development based on clearly defined metrics and project performance. A list of 20 performance measures was developed to assist in project selection and plan development. Sample Measure: • Reduction in hours of incident-related delay – the total incident delay for a given district in 2002. If a project reduces incident delay below the 2002 level, it only receives that portion of the improvement to the 2002 level. Table 4.2. Transportation Measures – Reliability

36and focus specifically on the congestion on those roadways that inhibit key travel for a particular population or trip type. Typical accessibility objectives include: • Provide residents access to regional centers; • Provide businesses access to market resources; and • Improve or maintain transportation equity. Table 4.3 lists five general measures that support these objectives, including examples of each measure. The case study highlight illustrates how Albany, New York’s congestion management process measures destination accessibility.Safety Safety refers to the ability for users of the system to reach their destination safely on any given trip. This is typically measured through the record of crashes or incidents along a particular roadway or at a specific intersection. Although transportation projects often focus exclusively on safety, the focus in this framework is on the safety impacts of highway capacity expansion projects. The following two measures (Table 4.4) support the objective of improving safety. The case study highlight demonstrates how the Denver Regional Council of Governments measures crash rates in their 2008-2013 Trans- portation Improvement Program.Table 4.3. Transportation Measures – Accessibility SHRP 2 Framework Measure Specific Measure Applications Job Accessibility – Number of jobs within a reasonable travel time for a region’s population. Destination Accessibility – Average travel time to major regional destinations. Labor Force Accessibility – Number of residents within reach of the region’s employers. • Percent of population within 30 miles of employment; and • Percent of population within 45 minutes of employment. • Average travel time from facility to destinations; • Origin-destination travel times; • Accessibility index; • Percent of population within five miles or 10 minutes of state-aided public roads; • Average number of job opportunities close (within 20 or 40 minutes, by peak auto- mobile and peak and off-peak transit); • Average number of home-based shopping opportunities (trips attracted by stores; based on 10-minute automobile and 20-minute transit travel times); • Average number of home-based other opportunities (within 20 minutes by auto- mobile and 40 minutes by transit); • Percent of population close to a college and close to a hospital (within 20 minutes by automobile and 40 minutes by transit); • Percent of population close to a retail destination (within 10 minutes by automobile and 20 minutes by transit); • Average travel time for work trips; • Average travel time for home-based shopping trips, home-based other trips; • Average travel time to the CBD; • Percentage of population group with transit access to the CBD; • Average number of jobs accessible within 15, 30, and 45 minutes by transit and automobile; • Average number of low-income jobs accessible within 30 minutes by transit; and • Average number of schools, food stores, health services, social services accessible within 30 minutes by transit and automobile. • Change in average travel time to major employment centers as result of project; • Change in number of employees within 45 minutes travel time to major employment centers as result of project; and • Percent of employers that cite difficulty in accessing desired labor supply due to transportation.

37Table 4.3. (Continued). SHRP 2 Framework Measure Specific Measure Applications Market Accessibility – Average travel time to market centers. Environmental Justice Accessibility Impact – Relative jobs, destinations, labor force, and market accessibility for environmental populations versus the general population. • Change in population within 45 minutes travel time to important market centers as result of project; • Percent of wholesale and retail sales in the significant economic centers served by unrestricted (10-ton) market artery routes; and • Percent of manufacturing industries within 30 miles of interstate or four-lane highway. • Level of access for disadvantaged populations to jobs, services, and market centers. Case Study Highlight: Albany, NY Congestion Management Process Description: Albany’s Capital District Transportation Committee uses their Congestion Management Process (CMP) to identify the regions’ conges- tion management needs as part of the region’s RTP. The CMP reports current values of performance measures and anticipated future values under alternative growth scenarios. These performance measures are related to transportation service (access, accessibility, congestion, flexibility), resource requirements (safety, energy, economic cost), and external effects (air quality, land use, environmental, economic). Sample Measure: • Travel Time between Representative Locations by Quickest Mode: – Sample Time: State Office Campus to Northway Exit 10 (minutes, P.M. Peak).SHRP 2 Framework Measure Specific Measure Applications Table 4.4. Transportation Measures – Safety Safety – Crashes per hundred million vehicle-miles traveled. Crashes – Absolute number of crashes over time (e.g., per year). • Accident risk index (‘safety index’); • Accidents (or injuries of fatalities)/PMT; • Fatality (or injury) rate of accidents; • Hazard index (calculated based on accidents per VMT by severity); and • Number of accidents per ton-mile traveled. • Accident rate, deaths, injury, property loss by type of corridor; • Average duration of accidents; • Number of pedestrian accidents (or injuries or fatalities); and • National rank for accident, injury, fatality rates. Case Study Highlight: Denver Regional Council of Governments FY 08-13 TIP Description: DRCOG’s project evaluation process for its latest Transportation Improvement Program (FY 2008-2013) includes a unique scoring system for each type of project, including roadway capacity. The scoring system is categorized into 10 topics: current congestion, safety, cost- effectiveness, condition of major structures, long range plan score, transportation system management, multimodal connectivity, matching funds, project-related Metro Vision implementation and strategic corridor focus, and sponsor-related Metro Vision implementation. Each cate- gory has a unique scoring system, and receives up to 15 points depending upon how that category is weighted. Project sponsors submit their project online, complete this ranking process, and are given an instant score. This gives them a sense of how their project will compare to others, and what areas they need to improve in order to increase the chances for funding. Sample Measure – Based on the project’s estimated crash reduction and weighted crash rate in comparison to the statewide average, up to 5 points will be awarded: • Using the estimated number of crashes reported by the applicant for the three-year period, the funding request application program will convert that to a per-mile basis and will assign the crash reduction level as follows: – Low (9 or fewer crashes eliminated per mile); – Medium (10-19); – High (20-29); and – Very High (30 or more).