Estimation of Annual Average Daily Traffic from one-week traffic counts. A combined ANN-Fuzzy approach

• A combined ANN-Fuzzy approach estimated AADT from one-week seasonal traffic counts.
• The accuracy of AADT estimates was satisfactory, according to FHWA requirements.
• AADT estimates were influenced by the period during which short counts were taken.
• The method is useful for planning of monitoring and minimizing traffic count costs.

This paper presents an approach to estimation of the Annual Average Daily Traffic (AADT) from a one-week seasonal traffic count (STC) of a road section. The proposed method uses fuzzy set theory to represent the fuzzy boundaries of road groups and neural networks to assign a road segment to one or more predefined road groups. The approach was tested with data obtained in the Province of Venice, Italy, for the period of the year in which STCs are taken. The method produced accurate results, which may be of interest for proper planning of monitoring and minimizing traffic count costs.