Towards efficient and accurate prediction of freeway accident severity using two-level fuzzy comprehensive evaluation

Area of Science:

• Intelligent transportation systems
• Road safety engineering
• Fuzzy logic applications

Background:

• Accurate freeway accident severity prediction is vital for effective road safety and emergency response.
• Existing methods lack the precision needed for intelligent freeway systems.
• Current limitations hinder timely and efficient freeway rescue operations.

Purpose of the Study:

• To develop an efficient and accurate Freeway Accident Severity Prediction (FASP) method.
• To categorize influencing factors into human (6 categories) and non-human (36 categories).
• To enhance the precision of freeway accident severity prediction.

Main Methods:

• Categorization of influencing factors into two hierarchical levels: human and non-human.
• Development of the Freeway Accident Severity Prediction (FASP) method using two-level fuzzy comprehensive evaluation.
• Utilizing fuzzy evaluation matrices, entropy method for weight calculation, and maximum membership principle for prediction.

Main Results:

• The FASP method demonstrates high accuracy in predicting freeway accident severity.
• Experiments on datasets from China and the US validate the method's effectiveness.
• The approach offers thorough consideration of factors with a low computational cost.

Conclusions:

• FASP provides an accurate and efficient solution for freeway accident severity prediction.
• This method represents a novel application of two-level fuzzy comprehensive evaluation in this domain.
• The findings contribute to improving intelligent freeway systems and emergency management.