The article reviews the studies carried out so far for road maintenance work. Exploring approaches in which raw data can be processed from road weather-stations to actual forecasting and forecasting model creation. The goal is to be able to make forecasting and build the best forecasting models that will be implemented into the BaSeCaaS platform in the future. Forecasting is designed to improve the current situation during the winter months for road maintenders for better decision-making. Initially, the missing data is filled to be able to make forecasting possible. Several methods are applied and identified, which is the best from an accuracy perspective. An experiment is conducted with ARIMA the best forecasting model for the particular dataset. As well as looking for the best approach to updating the forecasting model parameter to improve accuracy and better results. The concept is created under this article, and the BPMN of Road Maintainers Case process is reflected. Uptake of the current research is depicted in forecasting UML class diagram that is created and represented within the UML sequence diagram of the forecasting process.