Traffic demand is the basis for urban traffic planning and management. Most studies focus on exploring the temporal-spatial pattern of traffic demand. Few studies have investigated traffic demand structure with field data, such as the ratio of commuting demand. The availability of automatic license plate recognition (ALPR) data provides the opportunity to study the travel behavior of individual vehicles. A method was proposed to investigate the traffic demand structure of urban road networks using ALPR data. First, the temporal and spatial features of individual vehicles were extracted. A clustering technique was then used to identify the commuting vehicles. The commuting and non-commuting traffic demand of the road network were distinguished with the proposed method. Then the time-varying patterns of demand structure within a day combined with MFD and day-to-day patterns were analyzed. The proposed method would help conduct active, effective traffic management and control.
展开▼
