Trend forecasting for waste generation with structural break

For waste management planning, the adequate estimation of future waste production is crucial. Within waste production forecasting it is necessary to tackle with several challenges. The waste production data often consists of annual short time-series, but prediction horizon considers long-term estimation. Trend analysis seems to be a suitable approach for modelling waste production and its extrapolation. A methodology for forecasting waste generation based on historical data with a structural break is presented. To apply the approach, it is necessary to determine the estimation of separation potential. The principle is based on the idea of credibility theory where information from all territories is combined. Experience from other territories, which are more advanced in separation development, is used for forecasting. Finally, the population projection is taken into account to obtain information on absolute waste production. The case study is devoted to the forecasting of bio-waste production at the micro-regional level in the Czech Republic. There has been revealed an essential impact when population forecasting was included in the model. The increase in the average bio-waste production for micro-regions is expected. It will grow from 67 kg/cap/year in 2017 to 156 kg/cap/year in 2030. In 2030, most micro-regions would reach their potential based on this forecasting. •Waste production forecasting approach uses historical data with structural breaks.•Model based on the idea of credibility theory combines individual and general data.•Bio-waste data was analysed in the Czech Republic on the micro-regional level.•Bio-waste separation potential is estimated through building type of micro-regions.•The changes in population size rapidly affect the forecasts for waste generation.