Evaluation of the correspondence matrix by the Poisson method based on data from the online service for searching for car companions
For effective functioning of transport systems, information on spatial distribution of transport flows is necessary both at the current moment and for future periods. Traditional methods of obtaining observation results require very high costs, In addition, there is a high probability of obtaining low-quality data due to the strong influence of the human factor. The disadvantage of automated methods of traffic flow inspection is the high cost of equipment. The paper proposes a method for obtaining data on spatial distribution of passenger flows using constant monitoring of the online service of joint trips. To illustrate the possibilities of this approach, the online service for searching for car companions was used on the example of Sverdlovsk region. The correspondence matrix is constructed using a gravitational model applying the Poisson method. The Gauss - Seidlitz iterative method is used to find unknown parameters of the model. The assessment of quality and adequacy of the constructed model values of correspondence was carried out. It is shown that the spatial distribution of traffic flows obtained on the basis of data from online services for searching for car companions is well described by a gravitational model with a piecewise constant gravity function. Based on the analysis of the observed correspondence matrix, it was found that traffic flows between Yekaterinburg and all other cities of the Sverdlovsk region significantly exceed traffic flows between these cities.