In 2015, the United Nation member states developed and adopted a strategic blueprint aiming at transforming the world and ensuring prosperity for all by promoting peace and addressing contemporary challenges like poverty, energy, climate change and justice. This blueprint, dabbed the Sustainable Development Agenda of 2030, paved the way for the creation of 17 Sustainable Development Goals (SDGs), which call all countries to action regardless of their economic might. United Nations is committed to working with all governments and concerned stakeholders to push for the achievement of these SDGs. Importantly, environment and energy sustainability have been identified as crucial enabling factors in achieving and implementing the SDGs.
SDG 7 aims to ensure access to affordable, reliable, sustainable and modern energy for all. To achieve this goal, it is important to develop effective assessment models to evaluate the progress of different global economies towards affordable and clean energy. While there is good progress in classifying energy market based on well-defined development metrics and measuring the health and performance of energy services, this process is still challenging. This is further exacerbated by a lack of universal metrics and approaches for assessing the health and performance of energy markets. Popular metrics like energy poverty lack universality, while those based on advances in climate change experiences various drawbacks.
A holistic measure must include all the components of SDG7: affordable, reliable, sustainable and modern energy. However, most of the existing methods have mostly focused on the first two components. Sustainable energy is typically providing energy in a way that meets the needs of the current and future generations with little impact on the environment. It entails energy efficiency and renewable energy sources like biomass, solar, wind and hydroelectricity. Thus, any metric for evaluating energy markets and services that do not include sustainability and modernity is technically inadequate for assessing SGD7.
Herein, Dr. Zvikomborero Matenga from The Pennsylvania State University developed an empirical structure to effectively evaluate the advancement towards achieving SDG7 considering all four components. This holistic approach was based on machine learning using an unsupervised learning approach (ordinal K-Means clustering) to classify energy markets based on their performance and health. This model was adopted to track the growth and decline in energy markets considering three priority assignments: equality, accessibility and quality, using data obtained from the World Bank indicators between 1990 to 2019. The work is currently published in the peer-reviewed journal, Sustainable Energy Technologies and Assessments.
The author demonstrated the capability of machine learning in enabling a holistic approach to assessing the health of energy markets by combining numerous indicators into a single score. The use of the clustering approach enabled the definition of different health levels of markets considering the complexity of the factors involved. The obtained levels were more homogeneous than those obtained using other methods. The clustering method provided evidence of seven distinct ordinal classes that can be used for evaluating specific energy mares and developing effective energy policies. It also confirmed the poor performance of the Sub-Saharan energy markets in terms of accessibility and lack of significant progress towards sustainability in some developed economies like China.
Compared to previous approaches, unsupervised machine learning has numerous advantages, including the ability to learn and classify data without any labels, allowing the addition of the labels after data classification, useful in establishing patterns in data and allow effective evaluation of data changes over time using probabilistic methods to find the degree of data similarity. Furthermore, it will allow policymakers to customize needs analysis of a specific market and help developing countries to evaluate their markets in terms of reliability and affordability while tracking sustainability performance.
In summary, Dr. Matenga proposed a novel machine learning-based clustering approach to facilitate the assessment of the health and performance of energy markets toward achieving SDG7. The model was validated using myopic approaches. The finding singled out the dangers of myopic approach to effective evaluation of energy markets. In a statement to Advances in Engineering, Dr. Matenga stated that his proposed approach would allow policymakers to develop effective energy policies and evaluate their effectiveness and progress towards achieving SDG7.
Matenga, Z. (2022). Assessment of Energy Market’s progress towards Achieving Sustainable Development goal 7: A clustering approach. Sustainable Energy Technologies and Assessments, 52, 102224.