Social Energy System Design Incorporating AI and Lived Experience

2023 Fiscal Year Final Research Report

• Project/Area Number: 22K18039
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 63040:Environmental impact assessment-related
• Research Institution: Kyushu University
• Principal Investigator: Chapman Andrew 九州大学, カーボンニュートラル・エネルギー国際研究所, 准教授 (60795293)
• Project Period (FY): 2022-04-01 – 2024-03-31
• Keywords: energy system / preference / behavior / machine learning

Outline of Final Research Achievements

This research seeks to understand how people’s daily behaviors and preferences may influence their perceived importance of environmental, economic and social issues. To date a lot of research has been grounded in survey and statistical analysis-based approaches. Here, we seek determine the efficacy of decision tree machine learning approaches which only employ non-identifiable data to estimate people’s perceived issue importance based predominantly on behavioral inputs. Machine learning approaches as proposed in our framework can make predictions as to whether certain issues are important to people based not only on demographics but also on a suite of daily behaviors. This framework may provide a streamlined policy instrument for policymakers to develop energy policies which align with people’s values and therefore may be more effective for energy system design. In this research project we submitted two journal articles, 1 published and 1 is under review.

Free Research Field

Energy Analysis

Academic Significance and Societal Importance of the Research Achievements

The research is scientifically significant as it allows us to streamline the acquisition of data and it's application to machine learning to identify factors and preferences that were either unclear, or unable to be extracted from small data sets. Energy system design applications are also exciting.