• Federal Highway Administration >
• Publications >
• Research Publications >
• 07045 >
• Out.Cfm
• Applications of ISAT, Interchange Safety Analysis Tool (ISAT): User Manual

Applications of ISAT, Interchange Safety Analysis Tool (ISAT): User Manual

PDF files can be viewed with the Acrobat® Reader®

Output Reports

When the user clicks the Perform Calculations button on the Input-General worksheet, a number of output reports are generated within ISAT. ISAT automatically takes the user to the Output-General worksheet where output results are summarized in the following manner:

• Number of predicted crashes for entire interchange area.
• Number of predicted crashes by interchange element type.
• Number of predicted crashes by year.
• Number of predicted crashes by collision type.

Separate output worksheets are also generated for each type of interchange element (i.e., mainline freeway segments, ramps, crossroad ramp terminals and intersections, and crossroad segments). On these output reports the number of predicted crashes by collision type and severity level is summed across the individual components of the respective interchange elements, and results are provided for the individual components of the respective interchange elements.

Each of the output reports/worksheets is discussed in more detail below, using the output reports generated by the example that was illustrated throughout the user manual.

General Output Report

Five tables are included on the general output report (i.e., Output-General worksheet), titled:

Table 39 (General Interchange Information) reviews the input data from the Input-General worksheet. This table displays the project description, analyst, date, area type, and beginning and ending years of the analysis period. This example was of a rural diamond interchange, and the analysis period is from 2008 to 2017, inclusive. The table also displays whether crash data were incorporated into the calculations for the respective interchange elements. In this example, 5 years of crash data from 2001 to 2005 were considered in the calculations for each of the four interchange elements.

Table 40 (Number of Predicted Crashes for Entire Interchange Area) shows the number of predicted crashes during the analysis period for the entire analysis area and shows the average number of predicted crashes per year during the analysis period. These numbers are provided for TOT, FI, and PDO severity levels. For instance in the example, 210.2 TOT crashes are predicted to occur during the ten year analysis period within the entire interchange area, and this translates to 21.0 TOT crashes per year.

Table 41 (Number of predicted crashes by interchange element type) summarizes the crash predictions by interchange element types. This table shows the number of sites included in the analysis area for the four interchange elements. The table also shows the number of crashes by severity type that are expected to occur on each type of interchange element. Million vehicle-miles traveled (MVMT) are displayed for mainline freeway segments, ramps, and crossroad segments based upon the traffic volumes and lengths of these interchange elements. For crossroad ramp terminals and intersections, million entering vehicles (MEV) are displayed. The last column of the table displays the crash rates across each of the interchange elements, either based upon MVMT or MEV. The last row of the table shows totals for the entire interchange area. The crash rate in this last row is calculated using total crashes and total MVMT. In the example, 22 total sites are included in the analysis area (i.e., 10 mainline freeway segments, 4 ramps, 2 crossroad ramp terminals, and 6 crossroad segments). On the 10 mainline freeway segments, 109.9 TOT crashes, 29.0 FI crashes, and 80.9 PDO crashes are predicted to occur during the analysis period. The total MVMT across these 10 mainline freeway segments during the analysis period is 112.262. The average crash rate across the 10 mainline freeway segments is 0.979 TOT crashes/MVMT. The TOT number of crashes predicted to occur at the crossroad ramp terminals is 31.9 crashes. During the analysis period, it is calculated that 41.098 million vehicles will enter these two crossroad ramp terminals. This translates into an average crash rate for the two crossroad ramp terminals of 0.776 TOT crashes/MEV for the analysis period. The average crash rate across the entire interchange area during the analysis period is 1.541 TOT crashes/MVMT.

Table 42 (Number of predicted crashes by year) displays predicted number of crashes for each year of the analysis period by severity level. Predicted crash totals for the entire analysis period are also summarized by severity level. In the example, 19.7 TOT crashes, 5.5 FI crashes, and 14.2 PDO crashes are predicted to occur during the first year of the analysis period (i.e., 2008). During 2009, 20.0 TOT crashes, 5.6 FI crashes, and 14.4 PDO crashes are predicted to occur, and so on.

Table 43 (Number of predicted crashes by collision type) shows the number and percentage of predicted crashes by collision type and severity level. Subtotals are also provided for all single-vehicle crashes and all multiple-vehicle crashes. In the example, 210.2 TOT crashes are predicted to occur within the analysis area during the 10 year analysis period. Of these 210.2 TOT crashes, 129.3 crashes (61.5 percent) are expected to involve single vehicles, and 80.9 crashes (38.5 percent) are expected to involve multiple vehicles. Of the single vehicle crashes, 46.8 crashes are expected to involve a vehicle striking a fixed object.

Individual Output Reports for Interchange Elements

ISAT also generates output reports for each of the four interchange elements (i.e., mainline freeway segments, ramps, crossroad ramp terminals and intersections, and crossroad segments). Each of these individual reports is formatted in a similar fashion with two tables. The first table shows the number and percentage of crashes by collision type across all sites of the respective interchange element. These numbers and percentages are provided by severity level. The second table provides the predicted number of crashes for the individual sites for the respective interchange element along with additional descriptive information.

Tables 44 and 45 show the output results for mainline freeway segments. Table 44 (Number of predicted crashes by collision type for all mainline freeway segments combined) shows that 109.9 TOT crashes, 29.0 FI crashes, and 80.9 PDO crashes are predicted to occur across all the mainline freeway segments included in the analysis area. These crashes are further categorized by the 13 collision types. Table 45 (Number of predicted crashes for individual mainline freeway segments) shows the segment number, segment description, direction of travel, beginning milepost, ending milepost, and segment length for each individual site. For each individual site, the number of predicted crashes during the analysis period is provided by severity level. The average daily traffic (ADT) for each site is given for the entire analysis period along with MVMT, crashes per mile per year, and crash rate per MVMT.

As illustrated in table 45, several columns are provided to serve as warnings to users when interpreting the output. One warning is associated with the ADT of a given site. As indicated earlier, SPFs are used in predicting the number of crashes at a given site. Each SPF was calibrated/calculated using data from actual sites. The maximum ADT used in calibrating/calculating each SPF is one of the parameters included on the SPF worksheets. With a slight tolerance (i.e., 130 percent), if the ADT at a given site for any year in the analysis period exceeds the maximum ADT used to calibrate the SPF, which is used in predicting crashes for the respective site, this is indicted to the user by showing a "YES" in the respective row under the column headed "Max ADT for SPF exceeded". A "YES" is also provided in the first row of the table under the same column. These "YES" values serve as a warning to the user to view the predictions for the given site with caution. Because the ADT for the given site is beyond the limits for which the SPF was calibrated, there is less certainty associated with the predictions for the given site. A second warning is associated with the crash distributions. For a given group of crash types (i.e., 13 crash types) and subtype, the sum of the proportions should equal 1.00. If the sum of the proportions does not equal exactly 1.00, erroneous results will be calculated for the collision types. A "YES" value in the respective row under the column headed "Incorrect collision distribution" indicates that crash distributions used in the calculations for the given site were incorrect (i.e., did not sum to 1.00), and the user should adjust the respective crash distributions and re-run the analysis. Finally, the top row of this second table provides totals across all mainline freeway segment sites included in the analysis area.

Turning to the example, Segment No. 1 [EB Upstream Segment (MF1)] is 1.61 km (1.00 mi) in length. It is predicted that 14.9 TOT crashes, 3.7 FI crashes, and 11.2 PDO crashes will occur on this segment during the analysis period. The average ADT for this site during the 10 year analysis period is 4,741 veh/day. The total vehicle-miles traveled across this segment is calculated to be 17.304 MVMT. The average number of crashes per mile per year for this site is predicted to be 0.925 TOT crashes/km/yr (1.489 TOT crashes/mi/yr), and the crash rate is 0.861 TOT crashes/MVMT. No warnings are indicated for this site associated with the yearly ADTs exceeding the max ADT used to calibrate the respective SPF, nor for incorrect collision distributions. Results for the other individual segments can be interpreted in the same manner. As a reminder, the crash estimates for mainline freeway segments within interchange areas include crashes along the mainline freeway lanes, adjacent speed-change lanes (i.e., deceleration and acceleration lanes), and adjacent roadsides. These estimates are adjusted to account for acceleration lane lengths as input by the user. Crash estimates for mainline freeway segments outside interchange areas include crashes along the mainline freeway lanes and adjacent roadsides.

Tables 46 and 47 show the output results for ramps. The output results are displayed very similarly to those for mainline freeway segments with the following exception; there is no column for crashes per mile per year. In the example, there are 4 ramps included in the analysis area. The total length of these ramps is 2.09 km (1.300 mi). It is predicted that 13.6 TOT crashes, 2.7 FI crashes, and 10.9 PDO crashes will occur on these ramps during the analysis period. The average ADT across all of the ramps is 889 veh/day. The total vehicle-distance traveled across these four ramps is calculated to be 6.788 million vehicle kilometers traveled (4.218 MVMT), and the crash rate is 3.233 TOT crashes/MVMT. No warnings are indicated for any of the ramp sites associated with the yearly ADTs exceeding the max ADT used to calibrate the respective SPFs, nor for incorrect collision distributions.

Tables 48 and 49 show the output results for crossroad ramp terminals and intersections. The primary difference associated with these output results is that rather than reporting MVMT, exposure is represented by million entering vehicles (MEV). In the example, 2 crossroad ramp terminals are included in the analysis area. Focusing on the South Ramp Terminal (RT1), 13.4 TOT crashes, 6.2 FI crashes, and 7.1 PDO crashes are predicted to occur at this ramp terminal during the analysis period. A total of 19.467 million vehicles are expected to enter this ramp terminal during the analysis period, and it is predicted that 1.337 TOT crashes/yr will occur with a crash rate of 0.687 TOT crashes/MEV. No warnings are indicated for the site associated with the yearly ADTs exceeding the max ADT used to calibrate the respective SPF, nor for incorrect collision distributions.

Tables 50 and 51 show the output results for arterial crossroad segments. These output results are displayed exactly the same way as results for mainline freeway segments and would be interpreted in the same manner.

Previous | Table of Contents | Next