Predicting energy efficiency and carbon footprint in livestock farms using artificial neural networks: A case study in Turkey
Energy, cilt.365, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 365
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.energy.2026.142464
- Dergi Adı: Energy
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Applied Science & Technology Source, Compendex, Environment Index, Geobase, INSPEC, Public Affairs Index, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO)
- Anahtar Kelimeler: Dairy cattle farms, Energy balance, Environmental sustainability, Greenhouse gas emissions, Machine learning
- Bursa Uludağ Üniversitesi Adresli: Evet
Özet
On dairy farms, the increasing number of operational processes over time, along with rising energy costs and environmental issues, has made managing energy and its carbon footprint (CF) a necessity for farm sustainability. Thus, energy efficiency and energy-related CF were calculated for 93 dairy cattle farms in Türkiye, with capacities ranging from 1 to 225 heads, using conventional production methods and semi-open free-stall systems. Within the scope of the study, average energy consumption and output amounts, as well as energy efficiencies were determined using six parameters (feed, electricity, labor, machinery, young dairy cattle, fuel) and three output parameters (manure, milk, meat) obtained from field studies conducted on 93 farms. Additionally, average CF calculations based on energy sources have been carried out using the directly consumed energy inputs (electricity, fuel) and the inputs that drive energy consumption in production processes (feed and manure) for all farms. Finally, the energy efficiency and CF amounts for these farms were predicted using an artificial neural network (ANN). The average energy consumption of all farms was determined to be 78737.94 MJ/LSU. The average energy output from the farms was 13082.23 MJ/LSU. Accordingly, only 17% of the energy consumed in farms can be met. The average CF per liter of milk was 0.558 kg CO2-eq. On farms, the highest energy-related CO2-eq was fuel, while the lowest was manure. Using ANN, energy efficiency and energy-related CF have been successfully predicted, with R2 scores over 90% and low error rates. This will inform farms by providing a practical decision-support tool.