Sustainable Campus Policy Strategy in Estimating CO2 Emissions at the Universitas Negeri Semarang, Indonesia

In the fight against global warming, various options for reducing CO2 emissions are being implemented on campus. Furthermore, the management of campus sustainability at the Universitas Negeri Semarang (UNNES), Central Java, Indonesia, should be supported by accurate forecasts of electrical energy consumption. Therefore, this research aims to develop a predictive model to forecast the consumption of electrical energy in reducing CO2 emissions and to determine the factors triggering the increase. The prediction model is developed using Back Propagation Neural Network Artificial (BP-ANN) architecture. Furthermore, the data on the occupancy of lecturers and education staff as well as on students was obtained from the University's staffing and student affairs bureau. Climatic data such as temperature, humidity, wind speed, the duration of irradiation, and the average intensity of solar radiation were obtained per month from the Meteorology, Climatology, and Geophysics Agency of Semarang, Central Java for the 2013-2019 period as input data. The results of the empirical analysis showed an increase in electrical energy consumption from 2020 to 2025. In March, the consumption decreased but increased from April to June and decreased in July. It then increased until November and December, and it decreased every year. The results of CO2 emissions calculated by considering the emission factors from Indonesia's RUPTL-PLN in 2020-2025 showed an increase in electrical energy consumption and the ecological consequences affecting the campus area. Furthermore, the main factors causing the high consumption of electrical energy are the occupancy rate, lecturers, students, and campus employees, as well as local climate influences such as temperature, humidity, wind speed, duration of solar radiation, and intensity of solar radiation. Therefore, developing guidelines to reduce power consumption on campus should be a priority