Estimating truck travel patterns in urban areas

A method for estimating multi-class truck trip matrices from partial and fragmentary observations is presented. Data sets of widely varying character are combined in an efficient and effective manner so that each piece of information plays a role in developing the estimated flows. The method is linked to a geographic information system environment for data management and display of the results. Its use is illustrated through a case study focusing on the Bronx in New York City. Trip matrices are estimated for three truck classes: vans and medium and heavy trucks. Future advances for the method are outlined.