Estimation of Origin-Destination Matrix Using a Heuristic Algorithm

The purpose of this study is to propose an algorithm able to handle various constrains and network information while estimation origin-destination matrix. Due to the ability of the proposed algorithm to incrementally estimate the origin-destination matrix, it is capable of regenerating values close to real volumes of counted links. Furthermore, in order to obtain a unique matrix, some constrains in framework of an optimization problem are considered. These constrains include the confinement of the estimated matrix to the initial one as well as the confinement of the matrix elements to generation and absorption values in each zone in the target year. Finally, the performance of the proposed algorithm is evaluated through testing it on the Sioux Falls transportation network. The results demonstrated that the proposed algorithm is highly potential to estimate the origin-destination matrix in medium to large scale congested networks.