Home / The University / Faculties / Faculty of Economics and Business Administration / News / A FEBA team of scientists examined the possibilities of on-demand urban transport travel time prediction




Within the "INNOAIR" project, a team of The Faculty of Economics and Business Administration of Sofia University “St. Kliment Ohridski” published a scientific article examining the approach to predict travel time for on-demand public transport. The authors Angel Marchev, Boyan Lomev and Kaloyan Haralampiev present four predictive models, each of which is tested with data from specific scenarios.

The procedure for implementing the prediction model goes through three main phases: model identification, parameter estimation/optimization, and prediction with their respective subphases. Within their work, the researchers chose to use the following models to forecast the target data: Seasonal autoregressive moving average models with exogenous input with Fourier terms; Vector autoregression and vector error correction models; Bayesian Fourier model; Backpropagation neural network models. Root-mean-squared error, measured in seconds, calculated on the test set forecast (the last week of observations) is used to compare the proposed models.

The full text of the article can be accessed here.

innoair 2