PubAg

Main content area

Estimation of realistic renewable and non-renewable energy use targets for livestock production systems utilising an artificial neural network method: A step towards livestock sustainability

Author:
Elahi, Ehsan, Weijun, Cui, Jha, Sunil Kumar, Zhang, Huiming
Source:
Energy 2019 v.183 pp. 191-204
ISSN:
0360-5442
Subject:
buffaloes, economic sustainability, electricity, energy efficiency, equipment, farmers, farms, labor, livestock production, milk yield, millets, minerals, neural networks, nonrenewable resources, production functions, production technology, questionnaires, renewable energy sources, wheat straw, Pakistan
Abstract:
This study estimated energy use flow of buffalo farms, energy use indices, production efficiency, energy use targets, impact of energy inputs on energy output, and sensitivity analysis of energy inputs. A well-structured questionnaire was used to collect data of 360 domestic buffalo farms from Punjab Pakistan during May–July 2017. Results revealed that milk production was mainly dependent on renewable energy inputs, particularly millet, minerals, concentrates, and sorghum. Energy use efficiency (0.08) and production efficiency (0.24) indicate that energy inputs were overused. An artificial neural network (ANN) method suggested that 30.5% of total energy input could be saved if farmers followed the targeted inputs recommended by ANN. The Cobb-Douglas production function found a negative significant impact of sorghum, millet, and wheat straw; and positive significant impact of labour, concentrates, and electricity on energy output. Among the non-renewable energy sources, electricity was found to be the most wasteful use of energy input, mainly due to the mismanagement of farmers. Sensitivity analysis estimated that a unit increase in renewable energy significantly decreased milk yield by 0.02 unit. While a unit increase in non-renewable energy significantly increased milk yield by 0.01 unit. This study stresses the importance of using energy inputs at the target quantities prognosticated by ANN method, and recommends the use of energy-efficient equipment.
Agid:
6495261