Time series forecasting on cooling degree-days (CDD) using SARIMA model

Cooling Degree Days (CDD) is a convenient technique obtained by summing the cumulative air temperature differences that show how much deviation from the temperature is required for human comfort in a summer season. It is a basic and relatively simple measure for predicting the cooling energy requirements of the buildings. Accurate estimation of the seasonal trend of the CDD values is a crucial policy tool in determining the energy request for cooling the buildings and is critical to better energy management by decision-makers in the country. In this regard, planners or users need to develop appropriate and precise methods that allow them to forecast their future values based on their CDD historical time series data. For this reason, in this study, Seasonal Autoregressive Integrated Moving Average (SARIMA) model is utilized to forecast the CDD values in some regions with high cooling demand in Türkiye. To accomplish this, monthly CDD data from January 1991 to December 2022 are obtained from the provinces of Adana, Adyaman, Antalya, Siirt, and Şanlurfa. First, the CDD values are modeled using the SARIMA time series approach, and then the models are employed to predict the future trends of the CDD values from 2022 to 2031. Obtained results show that with the continuation of global warming at the current rate, CDD values in all selected provinces will increase slightly by 2031, which will cause a change in building energy consumption policies.