moved along each corridor, counting the number of crashes by density and severity by mode that occurred within each successive segment. An example of a sliding window analysis is shown in Figure 28 . The sliding windows score weights the most severe crashes more heavily than lower severity crashes. The analysis was conducted for each mode (bicycle, pedestrian, motorcycle, and motor vehicle), as
Figure 28. Example of the Sliding Window Analysis
well as CMV crashes. A crash was assigned a single mode based on the most vulnerable mode involved. For example, a crash between a motor vehicle and a bicyclist would be classified as a bicycle crash. For CMV crashes, any crash that is flagged as CMV-involved in CRIS (regardless of modes involved) was counted for that sliding windows analysis. The score for each window was determined based on the frequency and severity of crashes by mode. Fatal injury (K) and suspected serious injury crashes (A) were weighted x3, suspected serious injury (B) crashes were weighted x2, and possible injury (C) were weighted x1; no apparent injury (O), and unknown injury (U) were weighted x0. Once the weights are established and applied to the crashes, the number of crashes is aggregated to each window, incorporating the crash severity weighting. For example, if a segment had 1 A crash, 1 B crash, 2 C crashes, and 5 O crashes, it would receive a score of 7, or(1x3) + (1x2) + (2x1) + (5x0). High Injury Network Development Development of the HIN is an iterative process and should emphasize that the key goal of the safety action plan is the elimination of fatal and serious injury crashes. The sliding windows analysis helps to achieve that goal by providing scores for all segments. The next step is to
47
Made with FlippingBook - Share PDF online