Multifractal and singularity analysis of highway volume data

• The traffic data on work days and rest days have a similar fluctuation.
• Hölder exponents among a day, a month, three months and a holiday’s data.
• The singularity exponent method is more precise than the Hölder exponent method.

Recent work has shown that the mathematics of multifractal can be used to provide a quantitative signature in many fields. In this paper, we investigate the traffic time series for volume data observed on Guangshen highway. Firstly, we find there exists a multifractal behavior in the traffic data, and the data on both work days and rest days have similar multifractality. Then, we study the singularity of these data. A singularity exponent method based on multifractal theory is proposed to extract or classify singular data, which is more precise and clear than the Hölder exponent method.