THE MACHINE LEARNING TECHNIQUE FOR FORECASTING THE SEASONAL TIME SERIES

Abstract

Time series with seasonal variability is widely used to describe processes in various
fields, such as trade, analysis of financial markets, forecasting of passenger air transportation,
and description of climatic changes. Recently, this approach has been widely used to describe
technological processes as well. In this regard, applying predictive models in control systems of
complex technical objects has become possible. Machine learning methods can be effectively
used to build predictive models of series of this type. In this case, only historical data accumulated
over several periods of seasonal observation is used as input data for constructing the
forecast. Knowledge of other parameters, as a rule, is not required. The article considers creating
a predictive time series model with seasonal variability, describing a technological process,
the inlet flow of a wastewater treatment plant being chosen as a model. The general methodology
of model building, requirements for the input data sets, and algorithms of preprocessing to
form samples used for model training and testing are described. Classical methods (SARIMA,
Holt-Winters Exponential Smoothing, ETS), as well as new algorithms (Facebook Prophet,
XGBoost, Long Short Term Memory), were used to build the predictive model. The implementation
of the algorithms is done in the Python language, and recommendations for the use of existing
libraries and functions of this language are given in the work. The comparative analysis of
the accuracy of the obtained models is given on the calculation of a set of statistical metri cs.
Analysis of methods performance is also carried out since the time it takes to create a model
and get a forecast plays an important role when running the model in real production conditions.
The best method for solving the set task for application in real-time control systems was
chosen based on the sum of estimates. In conclusion, recommendations for improving forecast
accuracy were given, and future research directions were outlined.