Analysis Tool for Design Treatments to Address Nonrecurrent Congestion: Annotated Graphical User’s Guide Version 2 (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.

SHRP 2 Project L07 Analysis Tool for Design Treatments to Address Nonrecurrent Congestion Annotated Graphical User’s Guide Version 2.0 1

INTRODUCTION The SHRP 2 Project L07 Analysis Tool is designed to analyze the effects of highway geometric design treatments on nonrecurrent congestion using a reliability framework. The tool is a VBA interface (shown below) overlaying a Microsoft- based Excel spreadsheet (Excel 2007 and above). The tool is designed to analyze a generally homogenous segment of a freeway (typically between successive interchanges). The tool allows the user to input data regarding site geometry, traffic demand, incident history, weather, special events, and work zones. Based on these data, the tool calculates base reliability conditions. The user can then analyze the effectiveness of a variety of treatments by providing fairly simple input data regarding the treatment effects and cost parameters. As outputs, the tool predicts cumulative travel time index (TTI) curves for each hour of the day, from which other reliability variables are computed and displayed. The tool also calculates cost-effectiveness by assigning monetary values to delay and reliability improvements and comparing these benefits to expected cost over the life of each treatment. Note: This Guide uses the term “time-slice” in several places. A time-slice, in this tool, is a single-hour portion of a 24- hour day, considered over an entire year (excluding weekends and holidays). For example, “the hour from 6:00 a.m. to 7:00 a.m. for every non-holiday weekday between January 1 and December 31 of this year” is a time-slice. In this context, one way to think of a time-slice is as an “hour-year.” This User Guide is pictorial and annotated. It displays most of the entry screens presented to the user by the tool, with descriptions of their respective meanings and usage. Much of this help content (and additional guidance) is also embedded in the tool via information buttons. The Guide is not exhaustive; it describes neither the underlying theory nor the research that led to the development of the tool. The tool should be used in conjunction with two companion documents: the Project L07 Final Report and Project L07 Treatment Guidebook. This tool is the first of its kind, and reliability analysis is still in its infancy. Therefore, this tool and its successors will become more sophisticated in the future. 2

Overview Navigating the Tool The tool was designed to interactively predict reliability variables as a function of site conditions and traffic data, and allows the testing of the effects of different design treatments on reliability. To be viewed correctly, the screen resolution must be set to at least 1440x900. For Windows 7, make sure that on the Control Panel, under “Appearance and Personalization” \ “Display” \ “Make text and other items larger or smaller,” the value is set to 100%. Macro security: Excel may prevent you from opening this file if your security settings are configured to prevent macro- enabled files. In Excel’s trust center, you must select ”Enable all macros”; in Excel 2010, you must additionally select “Trust access to the VBA object model.” The main interface includes three main sections, as shown below; Site Inputs, Treatment Data and Calculations, and Results. Each of these sections is described in more detail later in this document. Site Inputs Treatment Data and Calculations Results 3

File Management Navigating the Tool Although the L07 Analysis Tool is based in Excel, data files are stored externally. Excel is used as an interface and calculation engine, but the data are stored in files called projects that are saved with an “.L07” extension (using an XML file format). What Is a Project? A project is a single data file that can store multiple sets of tool inputs. These sets are called scenarios. The tool allows three options: • New Project. Creates a new .L07 file, and populates the tool’s inputs with defaults. The user specifies the file name and which folder it should be placed in. • Open Project. Opens an existing .L07 file and populates the tool’s inputs with the project’s first scenario. • Copy Project. Creates a new .L07 file from the currently open project, saves the open project, and presents the new project for user input. Note that there is no “Save Project” command. A project file is saved whenever data from a scenario is saved, via a command from the scenario menu. What Is a Scenario? A scenario is a single set of tool inputs that lead to a cost-benefit calculation. Multiple scenarios can be included in a single project. (A project can also contain only one scenario if desired.) Ways to use scenarios could include: • Multiple segments—contain several segments of the same freeway for the same time horizon. (Example scenario names: “I-35_67th_to_75th,” “I- 35_75th_to_87th,” etc.) • Varying treatment characteristics—contain multiple options for the same treatment for the same segment of highway. For example, for the “Emergency Access Between Interchanges” treatment, the user may wish to analyze varying effects on emergency response time depending on treatment spacing. (Example scenario names: “reduction=5”, “reduction=10”) Essentially, the scenario concept allows the user to store dozens of “mini-files” (scenarios) in one single larger file (project). The user can toggle between scenarios at will using the “Current Scenario” drop-down. Scenarios can be added to a project, deleted from a project, copied within a project (“Save As”), or renamed. Advanced Feature: Merge Demand Data from External File A separate spreadsheet (L07DemandGen) has been created to allow the user to generate multiple demand scenarios (volumes and heavy vehicle percentages) without having to manually enter each of the 24 hours using the tool interface. The spreadsheet exports “.L07v” files that the user can then import into the tool using the “Merge Demand Data” item on the “Scenario” menu. The L07DemandGen spreadsheet includes an “Instructions” tab that serves as a user guide. Project Scenario 1 Scenario 2 Project Scenario 1 Scenario 2 Scenario 3 Adding a scenario Deleting a scenario Project Scenario 1 Scenario 2 Project Scenario 1 Scenario 2 Scenario 3 Copying a scenario (“Save As”) Project Scenario 1 Scenario 2 Scenario 3 Project Scenario 1 Scenario 2 Renaming a scenario Project Scenario 1 New Name Project Scenario 1 Scenario 2 4

Main Menu Bar Navigating the Tool The Main Menu Bar appears constantly at the top of the screen. It includes basic file operations, calculation settings, scenario management, print options, and basic help. Choose Automatic Calculation Selecting “Automatically Recalculate” directs the tool to automatically recalculate results any time an input is changed. Due to the complexity of the tool, calculation speeds may be slow, depending on the processing speed of the user’s computer. Therefore, when “Automatically Calculate” is deselected, the tool will calculate only when the user changes/moves to a new tab. Print Button The user has the option to print one or more of the available pre-designed printouts. The following window will pop up when the user hits the Print button: Project Name Help Button Close Button The “X” and Close buttons close the tool. The Help button provides an overview to guide the user through the steps of using the tool. The information found on the Help screen is similar to the information found in this User’s Guide. The user will then see a standard print pop-up, allowing selection of a printer and any relevant settings or options. Displays the name of the current project. Current Scenario Displays the name of the current scenario. The drop-down menu allows the user to select between the available scenarios for the current project. Project Selection Scenario Selection The project selection menu allows the user to start a new (blank) project, open an existing project, or make a copy of an existing project. Each project can contain multiple scenarios. The following window will pop up when the user hits the scenario button: Adds a new scenario to the current project Saves the data for the current scenario (and saves the project). Note that scenarios are also automatically saved when the user switches to another scenario. Creates a new scenario from the current scenario Changes the name of the current scenario Deletes the current scenario from the project Imports demand data from an external “L07v” file into the current scenario, replacing any demand data already included in the scenario (See the page on “File Management.”) 5

Information Buttons Information buttons are scattered throughout the tool. The user can click on them to see pop-ups providing guidance on what inputs mean, how inputs are used, what outputs mean, how treatments operate, and other items. Common Features Navigating the Tool Default Values In many cases, the input fields used by the tool are initially populated with default values developed as part of Project L07 research. The user may choose either to use these default values or to modify them for the analysis. When site-specific data are available, it is recommended that the user modify the default values to obtain the best possible results. To modify the defaults, the user activates the desired entry field by checking the small checkbox next to it. As can be seen in the example below, edited values will be red. Default Values Modified Values 6

Location This information appears as headers on the printouts but is not used in any calculations. Geometry These data are primarily used in the calculation of free- flow speeds (based on HCM procedures) for use in capacity/TTI calculations. Each of these terms is as defined in the 2010 HCM. Speed If “Measured FFS” is checked, the user can enter the field- measured value. Note that for HCM computations, this speed will be rounded to the nearest 5 mph (as the HCM 2010 requires). If measured speeds are unavailable, the user can check “Base FFS” and the tool will calculate the value based on the Geometry inputs above. Site Inputs Navigating the Screens On the Geometry tab, the user specifies basic information about the facility and provides basic geometric data allowing the tool to calculate speed and capacity. Geometry Base Lane Capacity By checking the box, the user may adjust the default lane capacity. Values are in passenger cars per hour per lane. If the box remains unchecked, the capacity is calculated based on the Speed values above. The capacity is used in d/c crit and other calculations. 7

Time of Day Beginning of time slice in 24-hour time. Demand These values are calculated using HCM 2010 procedures based on the user inputs on this tab. The demand flow rate is expressed as passenger car equivalents per hour (pcph). Heavy Vehicle Percentages Demand should be entered for the 30th- highest-hour of the year for the given time-slice. See the Project L07 Final Report for guidance on computing demand. Peak-Hour Factor Allows the conversion of hourly demand (D60) to peak 15- minute demand (D15). Currently unused by the tool. These are equivalent to the percent of each time-slice’s demand that is composed of trucks and recreational vehicles (RVs), respectively. Demand Flow Rate Demand Site Inputs Navigating the Screens On the Demand tab, the user provides information regarding traffic demand for each hour of the day. Typically, these values are collected in the field. The Project L07 Final Report contains more information on calculating demand. These values ultimately feed the calculation of the demand- to-capacity ratio, d/c crit, used in the TTI prediction models. Factor Volumes Clicking this button allows the user to adjust all 24 demand volumes up or down by a constant factor. To factor up, enter a positive percentage (e.g., 40% equals a 40% increase; all values will be multiplied by 1.4). To factor down, enter a negative percentage (e.g., -30% equals a 30% decrease; all values will be multiplied by 0.7). 8

Crashes Crash Costs Non-Crash Incidents Average incident costs by severity level are commonly available from state DOTs. The default values in the tool are based on national averages. Incident Site Inputs Navigating the Screens On this tab, the user provides information regarding both crashes and non-crash incidents. Data regarding non-crash incidents is not always available; therefore, the tool provides some default options to assist with incident estimation. Incident data are used in calculating lane-hours lost (LHL), a parameter in the TTI prediction functions. Crash costs and totals are also used in the cost-effectiveness calculations. Totals Enter annual crash totals by type (as typically recorded by highway agencies). The user can override the default average crash durations if better information is available. The user can also enter a percentage that indicates what portion of all incidents are crashes. This need only be used if, in the non-crash incident area below, the user has selected “calculate based on relation to crash %.” See the description at right for further details. If the user selects “Input number/year”, this section functions just like the Crashes section: the user enters the number of incidents/year, and can override the defaults for average duration if desired. In the absence of specific knowledge of non-crash incident totaIs, the user can select “Calculate based on relation to crash %”. The tool then supplies default values for the percent of each non-crash incident type with respect to all incidents (both crash and non- crash). In addition, above in the Crashes area, the tool supplies a default value for the percent of all incidents that are crashes. The user can override any of these defaults. Note that “Other” is automatically calculated so the total incidents sum to 100%. If the user enters a value that causes the total to exceed 100%, a pop-up will assist the user with rectifying the percentages. Disabled - Non-Lane Blocking refers to a car broken down on the shoulder, Disabled - Lane- Blocking refers to a car broken down in a travel lane, and Other refers to incidents not involving breakdowns, such as rubbernecking. Based on the information entered above, the tool computes and displays annual totals. 9

Precipitation Data Entry Options Option 2: Nearest Proxy Site Select one of these three options for entering weather (rain/snow) data for the site. Option 3: Specific Weather Data If the second radio button is selected above, the user can choose a specific weather station location (city) that is nearest to the treatment site. The tool will populate the “# of Hours” fields with precipitation data from that station. The location selected is displayed below the drop-down box. Weather Site Inputs Navigating the Screens Weather data are used in computing two precipitation variables used in the reliability prediction models: R0.05” (hours with rainfall exceeding 0.05 inches during the time-slice), and S0.01” (hours with snowfall exceeding 0.01 inches during the time-slice). The tool includes a built-in database of annual precipitation values at weather stations across the United States that can be used in the absence of detailed weather data. A 10-year average is used, which was collected from 2001–2010. The tool offers three options for entering precipitation data. If the third radio button is selected, the user enters precipitation values for each of the 24 time-slices. As stated above: In the Rain column, enter the annual number of hours with rainfall exceeding 0.05 inches, for each time-slice. In the Snow column, enter the annual number of hours (within each time-slice) with snowfall exceeding 0.01 inches, for each time- slice. For both rain and snow, the annual number should be normalized to a 250-day year of non- holiday weekdays. Option 1: Site Coordinates If the first radio button is selected above, the user can enter the latitude and longitude of the treatment site. The tool will select precipitation data from the nearest weather station and populate the “# of Hours” fields. 10

Number of Active Events An active (checked) event will be used by the tool in computations; an inactive (unchecked) event will not. These should be more broadly thought of as “event types” (e.g. baseball game, car race, etc.). Up to nine “event types” can be selected. Event Name Event Site Inputs Navigating the Screens The Event tab allows the user to enter data on special events that cause demand to fluctuate appreciably from the 30th-highest-hour values entered on the Demand tab. Events can include sporting events, concerts, etc. Based on frequency, event data are used to adjust the values entered in the Demand tab prior to using them in the reliability computations. The user may enter a name for each active (checked) event (e.g., “NFL Game”), for clarity. % Demand Increase For each active (checked) event, enter the percent demand increase caused by the event for each hour of the day. Example: for a baseball game, demand might increase 20% in the hour before the game, and 30% during the hour after the game. If the game typically starts at 7:00 p.m. and ends at 10:00 p.m., the user could enter “20” in the 18:00 box, “30” in the 22:00 box, and “0” in all other boxes. Event Frequency Enter the number of days per year that each active (checked) event occurs (or is expected to occur). Scroll Bar The user can scroll horizontally to access up to nine different events. 11

Active Work Zones An active (checked) short- term work zone will be used by the tool in computations; an inactive (unchecked) work zone will not. These should be more broadly thought of as “work zone types” (e.g., pothole repair, roadway widening, etc.). Up to nine “work zone types” can be selected. Work Zone Title Work Zones – Short Term Site Inputs Navigating the Screens The tool treats short-term (less than 30 days) work zones differently than long-term work-zones. The Work Zone tab allows the user to choose one or the other. For short-term work zones, the user enters data for work zones that reduce the capacity for the hours of the day when they are in place. These data are used in calculating lane-hours lost (LHL), a parameter used in the reliability prediction models. The user may enter a name for each active (checked) short-term work zone (e.g., “Widening”), for clarity. Lanes Closed For each short-term work zone, enter the number of mainline lanes closed for each hour of the day. For example, for a pothole repair that will close one lane from 4:00–6:00 a.m., the user enters “1” for Lanes Closed in the 4:00 and 5:00 boxes, and “0” for the remaining hours of the day. Enter the number of days per year that each active (checked) short-term work zone is in place. Note that short-term work zones can operate continuously or intermittently; the tool does not distinguish between the two. In other words, for the purposes of the tool, a single six-hour work zone is equivalent to two three-hour work zones. Scroll Bar The user can scroll horizontally to access up to nine different short-term work zones. Short vs. Long Term This button toggles between the long-term and short-term work zone analysis screens. Days Active Capacity per Lane The user may adjust the default capacity per lane of 1600 veh/hr by checking the box and entering an alternate value. 12

Work Zone Data It is expected that no more than one long-term work zone will occur at a given treatment site within a year; therefore, data for only one long- term work zone may be entered into the tool. The long-term analysis requires the user to enter many of the same inputs as the short-term analysis (length of work zone in days, number of lanes closed, and capacity per lane). In addition, the user should enter the lateral clearance (in feet) from edge of travel way to the work zone, the lane width (in feet), and the free-flow speed of the facility (in mph). Work Zones – Long Term Site Inputs Navigating the Screens The tool treats long-term (greater than 30 days) work zones differently than short-term work- zones. The Work Zone tab allows the user to choose one or the other. For long-term work zones, the user enters similar data defining capacity, but the work zone is treated as a “new normal” and is used as a base against which to compare treatments. If it is known that the long-term work zone will cause traffic to divert to other routes, enter the percent of the total demand that is diverted, for each hour of the day. Short vs. Long Term Diverted Traffic This button toggles between the long-term and short-term work zone analysis screens. 13

Demand Graph This graph displays the input demand and the calculated flow rate from the Demand tab. Graphs Site Inputs Navigating the Screens The Graphs tab is a display of the key input data as it varies by hour of the day, allowing the user to visually inspect the inputs from the other “Site Inputs” tabs. This graph displays the percent by which the volume at the site is increased for each type of active (checked) event. Incident Graph This graph displays daily distribution percentages for crash and non-crash incidents. The tool assumes that crashes are distributed in proportion to hourly density (calculated by the tool), and that non-crash incidents are distributed in proportion to hourly demand. More details on the distribution methodology are provided in the Project L07 Final Report. Events Graph Weather Graph This graph displays the total number of hours per year in each time-slice that experiences rain and snow exceeding trace amounts, as entered in the Weather tab. Short-Term Work Zone Graph This graph displays the percent by which the capacity of the segment is reduced for each active (checked) short-term work zone (from the Work Zone tab). 14

Title Bar Treatment Data and Calculations Navigating the Screens The treatment screens allow the user to test the effects of design treatments on site operations. The title bar contains elements that are common to the treatment analysis, allowing the user to select the treatments to be analyzed, the treatment to be viewed, and a description of the current treatment. The user can also enter default values for economic analysis. This button pops up a description of the currently selected treatment. Descriptions generally also contain photographs of the treatment and tips for analysis. Cost-Benefit Defaults This button pops up a window with default values used in cost-benefit analysis. The user can modify the value of time (VOT), reliability ratio (the ratio between the value of reliability (VOR) and the VOT), and discount rate. Treatment Description Treatment Name The name of the currently selected treatment is displayed here. Treatment Tabs This area of the screen includes one tab for each of the treatments to be analyzed for the site (selected from the Treatment List). Each tab is clickable and brings up its own individual screen where the user enters additional inputs (see following pages for input screens). Manage Treatment List This button pops up a list of treatments available for analysis. The user may select up to 10 treatments per site. For each treatment selected, a tab will appear in the Treatment Tabs. 15

Operational Inputs Treatment Data and Calculations Navigating the Screens For each non-custom treatment, the left column contains Operational Inputs. These vary by treatment, but a guide to the types the user will encounter is included below. Crash Data Enter the percent of each crash type that is expected to be affected by the treatment (in the way described for that treatment). Some treatments require additional information, as shown at right. Non-Crash Incident Data Enter the percent of each non- crash incident type expected to be affected by the treatment (in the way described for that treatment). Many treatments do not allow modification of all three types, because some treatments are not expected to affect some types of incidents. Other Data For certain treatments, additional inputs and application decisions are required. 16

Untreated Conditions Custom Raw Treatment Treatment Data and Calculations Navigating the Screens These fields are automatically calculated and filled based upon the data in the Site Inputs portion of the tool, and are provided for reference. Treated Conditions The user can manually enter/manipulate treated values. Essentially this allows the user to “create” a treatment if the user has knowledge of that treatment’s effects on the four key variables. This is also a way for the user to test base capacity improvements, ITS/operational strategies, and other improvements that might not fall under the category of “highway design to address non- recurrent congestion,” but that may have an impact on reliability. The TTI models are presented in the Project L07 final report, but can also be seen by clicking the Treatment Description button for this treatment. There are three custom treatment types available. For each of these, the left column looks quite different than what is shown on the typical treatment entry screen. The Custom Raw Treatment is the most complex of the three, requiring a deeper understanding of the TTI prediction models than the other two. It allows the user to directly manipulate the four model variables: lane-hours lost (LHL), hours of snowfall (S01), hours of rainfall (R01), and critical demand/capacity ratio (D/C crit). Thus, the effect of any treatment (or even condition) that can be mapped to these variables can be analyzed. 17

Untreated Conditions Custom Treatment - Flow Treatment Data and Calculations Navigating the Screens These fields are automatically calculated and filled based upon site inputs and are provided for reference. Treated Conditions The user can manually input treated demand and capacity values for each hour of the day. This allows the user to model any treatment or condition that is known to reduce demand or improve capacity. There are three custom treatment types available. For each of these, the left column looks quite different than what is shown on the typical treatment entry screen. The Custom Treatment-Flow tab allows the user to manually input the treated values for Capacity and Demand, thus allowing prediction of effectiveness for treatments that are known to influence either or both of these values. 18

Untreated Conditions Custom Treatment - Incidents Treatment Data and Calculations Navigating the Screens These fields are automatically calculated and filled based upon site inputs, and are provided for reference. Treated Conditions The user can manually input values based on treated conditions. There are three custom treatment types available. For each of these, the left column looks quite different than what is shown on the typical treatment entry screen. The Custom Treatment - Incidents tab allows the user to manually input the number of Crashes and Non-Crash Incidents expected under treated conditions, and to adjust durations. This allows the user to model any treatment or condition that is known to reduce crash rates or durations. 19

Cost Data Cost-Effectiveness Treatment Data and Calculations Navigating the Screens For all treatments (including custom treatments), the right column summarizes the cost-effectiveness analysis and has an identical layout for all treatments. The user is asked for several inputs related to treatment life, treatment costs, and treatment benefits. Enter the cost to construct the treatment, as well as the annual cost to maintain the treatment. The tool calculates a total annualized cost that is ultimately converted to a Net Present Value in the “Cost Effectiveness” section. Cost-Effectiveness This section of the tool converts the annualized costs and benefits to Net Present Values (using the USPWF), and then calculates the Net Present Benefit of the treatment by subtraction, and the Benefit-Cost (B/C) Ratio by division. Benefits The tool calculates the Annual Delay Reduction and Standard Deviation Change Indicator based on the operational improvements predicted for the treatment. From these, using the Value of Time and Value of Reliability, respectively, the tool calculates the Annual Operational Benefit. The Annual Safety Benefit is calculated based on the indirect effect of congestion reduction on crash rates, and on the direct effect of the treatment itself on reducing or eliminating certain types of crashes. Inputs Enter the expected service life of the treatment (how long the treatment will function before needing to be replaced, in years). The discount rate (essentially an annual interest rate that allows the computation of present economic value) can be user-modified (see Title Bar page). The Uniform Series Present Worth Factor (USPWF) is calculated by the tool and is used to convert an annual uniform cash flow to a present value. 20

Reliability Inputs Results Navigating the Screens To compute reliability results, the tool uses four variables: critical demand/capacity ratio (dc crit), lane- hours lost (LHL), hours with rainfall exceeding 0.05 inches (R0.05”), and hours with snowfall exceeding 0.01 inches (S0.01”). The Reliability Inputs tab includes graphs of these variables computed for each of the 24 time-slices, both in the untreated condition and with the treatment indicated at the top of the screen. For both conditions, the minimum and maximum values are indicated on the right side of each graph, along with the hours during which they occur. Critical Demand/ Capacity Ratio (dc crit) For the purposes of this tool, dc crit is the ratio of demand to capacity for a given hour. Number of Hours with Precipitation Exceeding Trace Values (R0.05” and S0.01”) R0.05” indicates how many hours experience 0.05 inches or more of rain during the time-slice. Similarly, S0.01” indicates how many hours experience 0.01 inches or more of snow during the time-slice. Lane-Hours Lost (LHL) LHL has two components: annual lane-hours lost due to incidents (ILHL) and annual lane-hours lost due to work zones (WZLHL). Untreated ILHL is based on data input on the Incident tab, and untreated WZLHL is based on short-term work-zone data input on the Work Zone tab. Treated values are based on the expected effects of the treatment. Refresh Graphs Name of treatment for which reliability inputs are being shown. Treatment Name If graphs disappear (a quirk of Excel), this button can be used redraw them. 21

Travel Time Index Results Navigating the Screens The Travel Time Index (TTI) is defined as the ratio of the actual travel time on a highway segment to the free-flow travel time. A segment operating at free-flow would have a TTI of 1.0; a segment on which cars traveled at half the free-flow speed (therefore taking twice as long to traverse the segment) would have a TTI of 2.0. The curvature of the graph is related to the reliability of a facility; the less the curve bends to the right, the more reliable the facility is. A perfectly vertical curve would represent a “perfectly reliable” facility. Percentile TTI Graphs These graphs represent the 24-hour graphs in a different form, with one curve for each of the five percentiles, for the untreated and treated conditions. 24-Hour TTI Graphs Refresh Graphs Name of treatment for which TTI graphs are being shown. Treatment Name If graphs disappear (a quirk of Excel), this button can be used redraw them. These graphs show TTI curves for all 24 time- slices, both in untreated and treated conditions. The tool calculates multiple TTI percentile values: 10th, 50th, 80th, 95th, and 99th. The values are connected to form the cumulative curve. The differences between the treated and untreated curves form the basis for the annual treatment benefit computations on the “Treatment Data and Calculations” tabs. Single-Hour TTI Graph This graph allows the user to select a single time-slice with which to compare treated and untreated TTI curves. 22

Reliability MOEs Results Navigating the Screens This tab displays graphs illustrating how several reliability measures of effectiveness (MOEs) vary by time of day, for both untreated and treated conditions. It also illustrates the difference between treated and untreated MOEs for each hour of the day. The MOEs are defined in more detail in the help menu. MOE Graphs Based on the user’s selection, the tool graphs the reliability measures for both untreated and treated conditions. Active Reliability Measures Refresh Graphs Name of treatment for which MOE graphs are being shown. Treatment Name If graphs disappear (a quirk of Excel), this button can be used redraw them. The user selects which reliability measures appear on the graphs below. Multiple measures can be graphed simultaneously. Difference Graph This graph subtracts the untreated MOE values from the treated MOE values for the selected measures. Positive values indicate an increase in the MOE value with the treatment; negative values indicate a decrease. 23

TTI Percentile Results Navigating the Screens This tab displays the TTI Percentile for each hour of the day in numerical format for both the treated and untreated conditions. This is the same data used to create the graphs on the “TTI” tab. Treated Results The mean TTI and five TTI percentiles for each of the 24 hours of the day for the treated condition. These data are graphed in the “Treated” graphs on the “TTI” tab. Untreated Results Refresh Graphs Name of treatment for which TTI Percentiles are being shown. Treatment Name If graphs disappear (a quirk of Excel), this button can be used redraw them. The mean TTI and five TTI percentiles for each of the 24 hours of the day for the untreated condition. These data are graphed in the “Untreated” graphs on the “TTI” tab. 24