Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results (2021) / Chapter Skim
Currently Skimming:
Pages 37-46
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 37... ...
37 Introduction The application of CPMs involves the use of SPF, optionally in combination with CMFs. Both the SPF and CMFs require several input values (e.g., segment AADT or lane width)
|
From page 38... ...
38 Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results Procedures to Assess Potential Reliability It is not possible to give strict guidance on how uncertainty in input values will affect the reliability of a CPM. This is because the impact on reliability can vary depending on (a)
|
From page 39... ...
Quantifying the Reliability of CPM Estimates for Error in Estimated Input Values 39 Step 2. Calibrate the CPM to Local Conditions Follow the HSM guidance (Part C, Appendix A)
|
From page 40... ...
40 Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results Step 6A. Estimate the RMSD, the Mean Absolute Difference, and the Extreme Value Use the values from Step 5 to estimate the mean difference, RMSD, the mean absolute difference, and the extreme value at the desired percentile.
|
From page 41... ...
Quantifying the Reliability of CPM Estimates for Error in Estimated Input Values 41 Step 8A. Assess the Impact of Measurement Errors on the CPM Using the GOF measures calculated in Step 7A, assess the impact of measurement errors on the CPM using the guidance in Table 8.
|
From page 42... ...
42 Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results Step 3. Assign a Random Number to Each Variable of Concern in the CPM The levels of measurement error assumed were 20% to 30% for AADT, 10% to 20% for dwydens, and 0% to 10% for medwid.
|
From page 43... ...
Quantifying the Reliability of CPM Estimates for Error in Estimated Input Values 43 Step 7A. Normalize the Measures Estimated in Step 6A The RMSD and extreme value were divided by the average predicted value per year using the CPM with known values, 1.33 crashes/year, and multiplied by 100 to express as a percentage.
|
From page 44... ...
44 Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results Step 1 through Step 5 Step 1 through Step 5 as described in Scenario 2, Case A, are the same for this scenario. At this point, the EB method is applied to combine the CPM prediction with the observed crash frequency.
|
From page 45... ...
Quantifying the Reliability of CPM Estimates for Error in Estimated Input Values 45 for each measure. The most reliable rating is High, while the worst is Critically Low.
|
From page 46... ...
46 Reliability of Crash Prediction Models: A Guide for Quantifying and Improving the Reliability of Model Results Step 9B. Tabulate the Top 30, 50, and 100 Sites For the top 30, 50, and 100 sites ranked using the base CPM, the percentage of sites not included (false positive sites)
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.