Short-term Safety Performance Functions by Random Parameters Negative Binomial-Lindley model for Part-time Shoulder...

Area of Science:

• Transportation Engineering
• Traffic Safety
• Roadway Operations

Background:

• Part-time shoulder use (PTSU) increases roadway capacity during peak hours.
• Limited quantitative data exists on PTSU design and operational impacts on freeway safety.
• Understanding crash factors related to PTSU is crucial for effective implementation.

Purpose of the Study:

• To analyze crash contributing factors associated with PTSU design and operation.
• To develop short-term Safety Performance Functions (SPFs) for PTSU segments.
• To compare the safety of PTSU sections versus non-PTSU freeway sections.

Main Methods:

• Development and comparison of statistical models, including Random Parameters Negative Binomial-Lindley (RPNB-L).
• Utilized mixed distributions and varying posterior parameter estimates for enhanced model performance.
• Analyzed crash data to identify significant contributing factors for PTSU implementation.

Main Results:

• The RPNB-L model demonstrated superior performance in analyzing crash data.
• PTSU implemented sections showed a lower frequency of traffic crashes compared to non-PTSU sections.
• Leftmost shoulder use, emergency rest areas, and adequate shoulder width significantly reduce crash frequency.

Conclusions:

• PTSU can be a safe and effective traffic management strategy when implemented with specific design considerations.
• Transition areas (start/end of PTSU) are critical hotspots requiring focused safety countermeasures.
• Findings provide valuable insights for transportation agencies to enhance safety on PTSU freeways.