Safety network screening for municipalities with incomplete traffic volume data

Area of Science:

• Transportation Engineering
• Road Safety Analysis
• Statistical Modeling

Background:

• Current road safety screening methods often require traffic volume data, which is frequently unavailable for many urban road segments.
• This data gap limits the effectiveness of traditional safety performance function (SPF)-based screening.
• Existing statistical tests like binomial and beta-binomial have been used, but without clear guidelines on when to apply each.

Purpose of the Study:

• To introduce a formal statistical test, the C(α) test, for determining the appropriate use of the beta-binomial test in road network screening.
• To enhance road safety analysis by providing a method applicable even when traffic volume data is missing.
• To aid decision-making for safety countermeasures by identifying high-risk road segments and collision types.

Main Methods:

• Development and application of the C(α) test for selecting between binomial and beta-binomial tests.
• Analysis of five years (2005-2009) of collision data for major arterial uncontrolled access segments in Saskatoon.
• Focus on rear-end and side-swipe same-direction collisions.
• Utilized ArcGIS for creating visual collision maps to aid decision-making.

Main Results:

• The C(α) test provides a formal statistical basis for choosing between the binomial and beta-binomial tests in network screening.
• Effectively identified road segments with high collision frequencies or proportions, particularly for rear-end and side-swipe collisions.
• Developed visually intuitive collision maps to support the implementation of safety countermeasures.

Conclusions:

• The C(α) test is a valuable tool for enhancing road safety network screening, especially in data-scarce environments.
• The study provides a more robust methodology for identifying hazardous road locations and specific collision configurations.
• Visual mapping tools can significantly improve the practical application of safety analysis findings for traffic management agencies.