Findings: This study included sixty-two (62) Hazard Elimination and Safety projects located throughout the state of South Dakota. Projects were located both in urban and rural areas. The roadways involved were highways and secondary roads. Of these sixty-two projects, there were seventeen (17) improvement types. Most of the improvement types included three (3) or four (4) project locations. Three (3) improvement types included only one project location. The largest improvement type, Traffic Signals, covers nine (9) project locations.
Decreases are shown in thirteen (13) improvement types with three (3) of these types showing a 100 percent accident reduction. Four (4) improvement types show accident increases. The increases in each of the four types could be attributed to one or two locations. No improvement type that showed an overall increase in accidents had all of its project locations increase in accidents. An example of this increase would be the improvement type, Reconstruction Left Turn Lane With Signal Phase, which has locations with ARFs of .61, .10, and 2.71. The final factor gives the overall ARF a percentage increase of 14 percent. The Accident Reduction Factor for the improvement type would have been close to .35 if the final factor was not included. This offset effect can be seen in most other improvement types that show an increase. One location’s figure throws off the composite result of the improvement type due to the lack of location data.
The accuracy of the results of this study increases with the number of projects studied. Results are more accurate for improvement types with a greater number of project locations. The improvement type Traffic Signals, which has nine (9) locations, is to be considered the most accurate. The improvement types Reconstruction-Increase Turning Radii, Remove Fixed Object, and Shoulder Widening are considered least accurate. Each of these improvement types are based on only one HES project. It is curious to note that the improvement type, Traffic Signals (comprised of 9 locations) is below the literature search average by more than 12 percent and that the improvement type, Shoulder Widening (comprised of one location) is equal to the literature search average. The lack of uniformity would reinforce the need to broaden the scope of each improvement type to include more locations in order to obtain more accurate results.
Accident Reduction Factors for twelve of the seventeen improvement types varied from the average of ARF’s found in the literature search by at least 10 percent. One South Dakota Average ARF, as mentioned earlier, is equal to its respective literature search average, two are within 5 percent and the remaining two are within 10 percent. The four improvement types with negative Accident Reduction Factors vary from the literature search averages from 49 percent to as much as 141 percent.
A Severity Reduction Ratio was calculated for each project location. A Severity Reduction Ratio would ideally be less than 1.00. This represents a lower overall severity of accidents after the completion of the improvement project. A SRR of 1.00 represents no change, and a SRR of greater than 1.00 would indicate an increase in severity. Out of sixty-two (62) individual HES project locations, twenty (20) showed a Severity Reduction Ratio greater than 1.00. Five (5) improvement types show overall SRR increases; twelve (12) locations have a SRR of 1.00 or less. As with the Accident Reduction Factors, overall Severity Reduction Ratio increases can be contributed to one or two outstanding project locations per improvement type. No improvement type that showed an overall increase in severity had all of its project locations increase in severity.
A cost/benefit analysis was performed on forty-eight (48) Hazard Elimination and Safety projects. These projects were funded solely by money set aside by the HES program. Cost/Benefits were found for individual project locations and for overall improvement types. Twenty-six (26) project locations were found to be beneficial both individually and overall. Eight (8) locations were neither benificial individually nor overall. Seven (7) projects were not beneficial individually, but were beneficial overall. Finally, five (5) HES projects were beneficial individually, but were not beneficial based on the overall improvement type. Fourteen (14) Hazard Elimination and Safety project locations were not analyzed due to project funding from non-HES sources.