Knowledge about future traffic in backbone optical networks may greatly improve a range of tasks that Communications Service Providers (CSPs) have to face. This work proposes a procedure for long-term traffic forecasting in optical networks. We formulate a long-terT traffic forecasting problem as an ordinal classification task. Due to the optical networks’ (and other network technologies’) characteristics, traffic forecasting has been realized by predicting future traffic levels rather than the exact traffic volume. We examine different machine learning (ML) algorithms and compare them with time series algorithms methods. To evaluate the developed ML models, we use a quality metric, which considers the network resource usage. Datasets used during research are based on real traffic patterns presented by Internet Exchange Point in Seattle. Our study shows that ML algorithms employed for long-term traffic forecasting problem obtain high values of quality metrics. Additionally, the final choice of the ML algorithm for the forecasting task should depend on CSPs expectations.

Modern IT and telecommunications technologies create new possibilities of data acquisition for the needs of traffic analyses and transport planning. At the same time, the current experience suggests that it is becoming increasingly difficult to obtain data on interurban travels of people in a traditional way (among others, in Poland there has been no comprehensive survey of drivers on the sections of non-urban roads since 2006). Within the framework of the INMOP 3 research project, an attempt was made to analyse the use of the Big Data application possibilities including data from SIM cards of the mobile telephony operator [1] and data from probe vehicle data (also known as “floating car data”), as data sources for carrying out the traffic analyses and modelling of travels by all means of transport in Poland. The article presents the manner, in which the data were used, as well as methodological recommendations for creating transport models at the national, regional and local levels. Especially the results of work can be applied for systematic passenger cars trip matrix update

The article presents a new approach to building a passenger rail traffic generation model. It uses data on the number of passengers at stations and railway stops obtained from the databases of operators on the rail transport market through the Office of Rail Transport – market regulator – combined with data on the model of the area around the station built based on population, number of beds, individual motorization and gross domestic product (GDP). This enabled analyzing the potential of railway traffic generation at a very detailed level. The article presents a methodology for building a passenger rail traffic generation model and verification of this model based on limited variables describing railway stations and stops as well as traffic zones and available statistical data. The model takes into account three segments of the railway market: regional, interregional and inter-agglomeration transport. The results of these analyzes can be used to increase the accuracy and the reliability of rail traffic models used in the analysis of transport networks.