چكيده به لاتين
To implement advanced traffic management, some information such as people’s movement patterns is needed. OD demand matrix approximately defines the movement pattern which is used in transportation planning and traffic management. Estimation of the OD demand matrix is one of the most crucial issues in traffic management needing a prior matrix and collected traffic data to be conducted. The collected traffic data, depending on the available data collection tools, can be traffic volume, density, or travel time. OD matrix estimation problem is a multi-dimensional problem being affected by various factors.
With this in mind, in this thesis, it has been tried to shed light on the effects of traffic data types on OD matrix estimation. To carry out the OD estimation, a grid network is designed, and the OD matrix is estimated by using Simultaneous perturbation stochastic approximation (SPSA). Traffic volume, travel time, and density are the traffic data types that are considered in this thesis. It is concluded that traffic volume has the most effect on the OD estimation, and also the most contribution to decreasing the estimated OD’s error. Besides, it is demonstrated that even with a few traffic volumes, it is possible to get the desired estimated OD matrix.
These results help traffic engineers to have a deep understand of the OD matrix estimation procedure, and collect the traffic data more wisely.
