Pilot Testing of SHRP 2 Reliability Data and Analytical Products: Washington (2014)

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.

5 CHAPTER 1 Introduction 1.1 General Background One of the purposes of the second Strategic Highway Research Program (SHRP 2) is to improve the reliability of highway travel times by reducing the effects of nonrecurrent traffic events, including traffic incidents, work zones, demand fluctuations, special events, traffic control devices, weather, and inadequate base capacity. The following five research projects in the SHRP 2 Reliability area have produced guidelines and analytical tools for travel time reliability measurement, monitoring, enhancement, and impact assessment to be tested in this project:  L02: Establishing Monitoring Programs for Travel Time Reliability;  L05: Incorporating Reliability Performance Measures into the Transportation Planning and Programming Process;  L07: Evaluation of Costs and Effectiveness of Highway Design Features to Improve Travel Time Reliability;  L08: Incorporation of Nonrecurrent Congestion Factors into Highway Capacity Manual Methods; and  C11: Development of Improved Economic Analysis Tools. Specifically, these projects aid in quantifying the travel time reliability characteristics, identifying possible solutions for reliability improvement, and analyzing the potential effects of implementing those solutions. The products from these five projects can be classified into three categories: travel time reliability measurement and monitoring (L02), analysis and impact assessment (L07, L08, and C11), and project prioritization (L05 and C11). SHRP 2 L02 developed a travel time reliability measurement system (TTRMS), along with a guide, that is intended to show practitioners how to develop such systems. The analytical tool produced by the SHRP 2 L07 project is used to evaluate the cost-effectiveness of geometric design treatments for reducing nonrecurring congestion. The Excel spreadsheet-based analytical tool has incorporated SHRP 2 L03 methods, such as before/after analysis and a cross-sectional statistical model (Cambridge Systematics 2010). This tool can assist in estimating operational effectiveness and economic benefits of a variety of design treatments for specific highway segments. SHRP 2 L08 developed a procedure to estimate travel time reliability and the impacts of nonrecurrent congestion factors in the highway capacity context. Two Excel spreadsheet tools, FREEVAL and STREETVAL, have been developed to evaluate the change in travel time reliability associated with a variety of traffic characteristics using a scenario generator for freeways and signalized roadways, respectively. SHRP 2 C03 developed a case study–based economic impacts estimation web tool called T-PICS. The new tool developed by the SHRP 2 C11 project is also an Excel spreadsheet-based tool, serving as an extension of the SHRP 2 C03

6 tool to enable a wider range economic analysis. The tool uses separate sketch methods to predict the incident-induced delay and combines with the recurring delay to obtain mean travel time index (TTI), which serves as the predictor variable to measure all types of variations. SHRP 2 L05 provides a guide with five steps for incorporating reliability into planning and programming in order to generate support for funding to improve reliability. The primary audience groups are managers and decision makers. It also includes a technical reference for practitioners that describes the tools and data needed (recipes) to calculate performance measures. Effective transportation is critical for maintaining Washington’s economy, environment, and quality of life. Therefore, WSDOT has long been promoting a reliable, responsible, and sustainable transportation system. WSDOT’s economic vitality and renowned livability plan also targets reliability improvement as the state’s primary transportation goal for planning, operations, and investment. “Moving Washington” is a proven approach as well as investment principle for creating an integrated, 21st-century transportation system. It is also the framework for making transparent, cost-effective decisions that keep people and goods moving and support a healthy economy, environment, and communities. The Puget Sound area in Washington State has several ideal sites for testing the SHRP 2 reliability research products. The various kinds of traffic data collected on the freeway and highway network in this area can be used for evaluating the analytical tools. Through this research project, the research team has made solid moves toward accomplishing the following objectives: (1) incorporate the analysis products into the business and decision-making process; (2) improve the capability of analyzing travel time reliability at facility, corridor, and network levels; and (3) test the validity and usability of the SHRP 2 reliability products. 1.2 Introduction of SHRP 2 Reliability Data and Analytical Products SHRP 2 L38 focuses on testing products from five research projects: SHRP 2 L02, L05, L07, L08, and C11. The following overview of these research project products introduces the main features of each product and the relevant specifications. 1.2.1 SHRP 2 L02: Establishing Monitoring Programs for Travel Time Reliability SHRP 2 L02 focuses on measuring reliability, identifying factors affecting systems’ reliability, and proposing solutions for reliability enhancement (Institute for Transportation Research and Education 2013). Products developed through this effort are summarized in Table 1.1.

7 Table 1.1. SHRP 2 L02 Reliability Product Summary Products 1. A guide and supporting methodologies; 2. Travel time reliability monitoring system (TTRMS); and 3. Approach on synthesizing route travel time distribution from segment travel time distributions. Research team North Carolina State University; Kittelson & Associates, Inc.; Berkeley Transportation Systems, Inc.; National Institute of Statistical Sciences; University of Utah, and Rensselaer Polytechnic Institute. Input 1. Infrastructure-based sources:  Loop detectors,  Video image processors,  Wireless magnetometer detectors, and  Radar detectors. 2. Vehicle-based sources:  Vehicle-based detectors collect data about specific vehicles, either when they pass by a fixed point (automated vehicle identification, or AVI, data) or as they travel along a path (automated vehicle location, or AVL, data).  Automated vehicle identification (AVI) data collection includes Bluetooth readers and license plate readers (LPRs), radio-frequency identification, vehicle signature matching data.  Automated vehicle location (AVL) data include data from the Global Positioning System, connected vehicles, and cellular telephone network. 3. Nonrecurring event data:  Incident, weather data, work zones, special events. Output 1. Segment travel time, including its distribution; 2. Route travel time, including its distribution; 3. Sources of unreliability; and 4. The impact of the sources of unreliability. Description The project team conducted five case studies using various data collection technologies to develop methods for assembling and visualizing travel time reliability information. Memo This work builds on data generated by current traffic monitoring systems to provide a long- term picture of travel time reliability. Test locations San Diego, California; Northern Virginia; Sacramento–Lake Tahoe, California; Atlanta, Georgia; and New York–New Jersey. Accuracy Accuracy may be limited by quality of data sets for travel times, weather, incidents, etc. Strength An agency that implements a TTRMS will understand much better the reliability performance of its systems and monitor how its reliability improves over time:  What is the distribution of travel times in their system?  How is the distribution affected by recurrent congestion and nonrecurring events?  How are freeways and arterials performing relative to performance targets set by the agency?  Are capacity investments and other improvements really necessary given the current distribution of travel times?  Are operational improvement actions and capacity investments improving the travel times and their reliability? Weakness  Not considered that nonrecurring events can have large variances in severity; and  Roadway improvements targeting reliability are more likely to happen at segment level than route level, but segment-level reliability analysis is not addressed.

8 1.2.2 SHRP 2 L05: Incorporating Reliability Performance Measures into the Transportation Planning and Programming Processes SHRP 2 L05 provides a concise description of how to incorporate reliability considerations into the transportation planning and programming process, with a focus on helping agencies make choices and tradeoffs about funding and project priority (Cambridge Systematics 2013). Overview of SHRP 2 L05 is summarized in Table 1.2. Table 1.2. SHRP 2 L05 Reliability Product Summary Products 1. The reference guide 2. The technical reference Research team Cambridge Systematics, Inc. Input  Reliability measure that the leadership, staff, and stakeholders understand and that yields consistent results;  Reliability benefits of each project in the project list; and  An approach to estimate the impact of a project on reliability, such as sketch planning method, model post-processing tools, simulation, and monitoring and management tools. Output A list of prioritized projects based on appropriately selected approaches. Description To develop the means—including technical procedures—for state DOTs and MPOs to fully integrate reliability performance measures and strategies into the transportation planning and programming processes. Memo For product 1, the audience is planning, programming, and operations managers who are responsible for making funding decisions at state DOTs and MPOs. For product 2, it is intended to support analysts who will be developing and applying the technical approach for measuring reliability and making choices and tradeoffs. Test locations Colorado DOT, Florida DOT, Knoxville, TN MPO, LAMTA (Los Angeles), NCTCOG (Dallas–Fort Worth), SEMCOG (Detroit), Washington State DOT. Accuracy Simulation method is the most accurate assessment. Strength 1. Sketch planning method: easy and fast, use generally available data; 2. Model post-processing tools: link-level data: more robust than 1, based on local data from the established regional model; 3. Simulation or multiresolution methods: provide most robust forecast of TTV, combining TDM provide most accurate assessment of long-short term impacts on reliability; and 4. Monitoring and management tools: easy and fast once system is developed, based on real-world data. Weakness 1. Sketch planning method: limited reliability metrics, apply to aggregated conditions; 2. Model post-processing tools: require a regional TDM, limited reliability metrics; 3. Simulation or multiresolution methods: requires regional TDM and simulation model be available; time and resource intensive; and 4. Monitoring and management tools: analysis capability limited by data availability and quality, cannot test future strategies to address congestion.

9 1.2.3 SHRP 2 L07: Evaluation of the Costs and Effectiveness of Highway Design Features to Improve Travel Time Reliability The objective of SHRP 2 L07 is to evaluate the cost-effectiveness of geometric design treatments, such as alternating shoulders, emergency pull-offs, etc., in reducing nonrecurrent congestion (Potts et al. 2013). Products of SHRP 2 L07 are summarized in Table 1.3. Table 1.3. L07 Reliability Product Summary Products Spreadsheet-based analysis tool. Research team Midwest Research Institute (MRI). Input 1.Treatments 2. Data: (1). Geometric data:  Number of lanes / lane width  Right/left shoulder width  Number of interchanges per mile (2). Traffic data:  Free-flow speed  Demand volume (by hour of day)  Peak hour factor (by hour of day)  Percentage of trucks (by hour of day) and percentage of RVs (by hour of day) (3). Crash statistics for roadway segment:  Total annual property damage only (PDO) crashes  Total annual minor-injury crashes  Total annual serious- and fatal-injury crashes (4). Information about typical crash duration (time until cleared):  Average crash duration (min) for PDO crashes  Average crash duration (min) for minor-injury crashes  Average crash duration (min) for serious- and fatal-injury crashes (5). Other:  Information about special events (e.g., number, percent increase in volume)  Information about work zones 3. Benefits and Costs Output Evaluation results of cost-effectiveness for a treatment, such as TTI, reliability measures of effectiveness (MOEs). Description What does the tool do?  Implements project L03 models  Computes cumulative TTI curve for untreated and treated conditions  Estimates traffic operational effectiveness of design treatments at specific locations  Compares economic benefits of various design treatments at specific locations Memo In addition to the defined treatments available for analysis in the tool, users are also able to evaluate any other treatment they wish, provided treatment’s effect on the three model variables can be ascertained. Test locations Seattle, Washington. Accuracy The tool tends to underestimate the vehicle travel time when traffic flow is high.

10 Strength The tool can be used to measure the operational effectiveness as well as the economic benefit of design treatments for a freeway segment of interest. The tool allows highway agencies to compare the benefits and costs of implementing various nonrecurrent congestion treatments at specific locations. Weakness  The tool interface is not very user friendly. It runs into crash sometimes.  Detailed output information is not applicable, which limits the tool usability. Table 1.4. L08 Reliability Products Summary Products 1. Guide describing travel time reliability concepts for HCM audience, provides step-by- step processes for predicting travel time reliability for freeway and urban street facilities, and illustrates example applications of the procedures. 2. FREEVAL and STREETVAL Computational Engine. Research team Kittelson & Associates, ITRE, Cambridge Systematics. Input Main source of travel time variability, given scenario (time of day, road condition, severity, etc.), demand, capacity. Output HCM performance measure, the impacts of variability on performance over a year. Description Determining how data and information on the impacts of differing causes of nonrecurrent congestion (incidents, weather, work zones, special events, etc.) in the context of highway capacity can be incorporated into the performance measure estimation procedures contained in the HCM. Memo The methodologies contained in the HCM for predicting delay, speed, queuing, and other performance measures for alternative highway designs are not currently sensitive to traffic management techniques and other operation/design measures for reducing nonrecurrent congestion. A further objective is to develop methodologies to predict travel time reliability on selected types of facilities and within corridors. Test locations Three locations were selected for testing in the Puget Sound Region: I-5, I-405, and SR 522. Accuracy STREETVAL: Large discrepancy between software output and ground truth data. FREEVAL: Software provides a reasonable estimation of the travel time reliability. Strength STREETVAL: Employs a powerful random scenario generation process that is a powerful method for accounting for all possible likely scenarios. FREEVAL: Tool is able to provide a reasonable estimate of the travel time reliability. This suggests that the principal factors affecting reliability have been accounted for. Weakness FREEVAL: Weather events with marginal impact are excluded; assume incident occurrence and traffic demand are independent of weather condition. STREETVAL: The methodology does not address the events (e.g., signal malfunction, railroad crossing, signal plan transition, and fog dust storms, smoke, high winds or sun glare). Overall: The power in a prediction model lies in the idea that with limited information, an outcome can be deduced. A major drawback of these tools is that they require a large quantity of input data before they are able to make their predictions (this is especially true of STREETVAL) and this makes these tools both difficult and costly to implement from a practitioner’s point of view. It begs the question of whether these tools be simplified, lessening the amount of input data requirements, and still give reasonable reliability estimates?

11 1.2.4 SHRP 2 C11: Development of Improved Economic Analysis Tools Based on Recommendations from SHRP 2 L03 SHRP 2 C11 provides a sketch-level planning tool based on SHRP 2 L03 research that estimates the benefits of improving travel time reliability for use in benefit–cost analysis (Economic Development Research Group 2013). The SHRP 2 C11 products are summarized in Table 1.5. Table 1.5. SHRP 2 C11 Reliability Product Summary Products 1. Analytical tools; and 2. User guide. Research team Economic Development Research Group, Cambridge Systematics. Input 1. Travel time reliability  Scenario data and traffic data  Time/travel cost and reliability ratio 2. Market access  Facility type, such as marine, freight rail, air passenger, air cargo, passenger rail, etc.  Roadway improvements 3. Intermodal connectivity  Impedance decay factor and impedance data  Productivity elasticity  Impact zones and activity data Output 1. Travel time reliability (result for base year and forecast year)  Congestion metrics  Total annual weekday delay (veh-hrs)  Total annual weekday congestion cost for passenger and commercial vehicles, respectively 2. Market access (result for project/policy baseline and alternative)  Accessible employment  Concentration index  Commuter costs  Effective density/potential access scores 3. Intermodal connectivity  Facility connectivity raw value  Value of time savings for facility  Weighted connectivity 4. Final result  Value of traditionally measured benefits and wider economic benefits in target year for passenger trips and commercial (freight delivery) trips, respectively. Description Development of improved economic analysis tools based on recommendations from project C03. Memo T-PICS is a web-based sketch planning tool that allows state departments of transportation (DOTs), metropolitan planning organizations (MPOs), and other agencies involved in highway capacity planning to quickly estimate the likely range of impacts of proposed projects. Test locations Uses the L03 Data Poor models as the basis. Accuracy As a sketch planning tool, it provides good enough accuracy.

12 Strength With minimal data input, the tool adds value by incorporating change in travel time reliability into project economic analyses. Weakness The calculation methodology is designed to capture the benefits of major capacity projects. It is not sensitive to the travel time reliability changes associated with improvements at roadway intersections, interchanges, and freeway ramps. 1.3 Research Objectives This research project has two major objectives:  To provide feedback to SHRP 2 on the applicability and usefulness of the products tested; and  To assist agencies in moving reliability into their business practices through testing of the products developed by the five SHRP 2 Reliability projects. For testing the SHRP 2 Reliability Data and Analytical Products, the research procedure consists of three major steps: (a) data compilation, integration, and quality control; (b) experiment design for testing different products; and (c) test results evaluation and possible improvements. The L38D research team has followed the proposed procedure through the pilot testing of all the committed research products. 1.4 Final Report Organization This report contains nine chapters. Chapter 1 introduces the general background for the SHRP 2 L38 project and summarizes the objectives of the research project. The general testing approach is presented in Chapter 2. Chapter 3 describes the data compilation and quality control process applied to the data used for this study. Chapters 4–8 provide the details of the research in analyzing reliability and improvement strategies, including site selection, case description, testing results, comparisons, and discussions of the L38 tools. Based on the testing results, Chapter 9 concludes the research and offers potential improvement directions for the tested SHRP 2 Reliability products.