Evaluating the influence of road lighting on traffic safety at accesses using an artificial neural network

Abstract

Objectives: This article focuses on the effect of road lighting on road safety at accesses to quantitatively analyze the relationship between road lighting and road safety.

Methods: An artificial neural network (ANN) was applied in this study. This method is one of the most popular machine learning methods and does not require any predefined assumptions. This method was applied using field data collected from 10 road segments in Nanjing, Jiangsu Province, China.

Results: The results show that the impact of road lighting on road safety at accesses is significant. In addition, road lighting has a greater influence when vehicle speeds are higher or the number of lanes is greater. A threshold illuminance was also found, and the results show that the safety level at accesses will become stable when reaching this value.

Conclusions: Improved illuminance can decrease the speed variation among vehicles and improve safety levels. In addition, high-grade roads need better illuminance at accesses. A threshold value can also be obtained based on related variables and used to develop scientific guidelines for traffic management organizations.

Keywords:

• Artificial neural network
• accesses
• road lighting
• road safety

Additional information

Funding

This study is sponsored by Philips Lighting Research China.