Causal modelling to quantitatively analyse people’s everyday decisions regarding energy consumption and their reactions to interventions
Cargando...
Fecha
2024-12-17
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Deusto
Resumen
The goal of this research is to mitigate the effects of climate change by refining energy systems modelling (ESM) to obtain a more nuanced description of demand drivers, with a specific focus on residential energy consumption. The approach involves several key strategies: applying novel causal models to dissect human decision-making processes in energy consumption, particularly in the areas of mobility, flexibility, building (heating and cooling), and appliance use; developing innovative methods for forecasting energy loads and assessing the impact of policy changes; and deepening understanding of household energy consumption patterns, with an emphasis on factors influencing behavioural changes toward energy-efficient practices.
The WHY project, which has generated essential information on factors affecting household investment decisions in the current energy transition context, is a central element of this effort. Using a scenario-based methodology adapted from the Delphi method, the project has gathered expertise on future energy transition scenarios in four key areas. Through a unique co-creation activity, based on self-determination theory, 32 determinants of investment decision-making were identified and systematically categorised into a taxonomy of 32 factors. This taxonomy was tested through a survey of more than 1,700 people in the EU and Latin America to determine how these determinants are distributed across the target population.
In addition, a cross-sectoral survey was conducted, validating archetypes and prioritising factors within the energy transition, using fictitious scenarios. This survey played a key role in characterising the archetypes of the European population and identifying behavioural clusters, thus enriching our understanding of investment archetypes. Advanced AI techniques, such as clustering and Monte Carlo simulations, were employed to discern deterministic clusters, leading to the emergence of new archetypes, such as "the activist". These clusters, representing various investment profiles, were further detailed through "person descriptions," which facilitated a deeper understanding of decision-making behaviours and underlying needs. The results, including the datasets, have been made public.
The final stage of this research was to generate causal models for each investment profile, using the stages of change of the Transtheoretical Model (TTM) to categorize each determinant. Directed Acyclic Graphs (DAGs) were used to conceptualise the causal relationships between the influencing factors, providing essential information for shaping energy policies. These DAGs enable targeted interventions on priority factors, thus streamlining decision-making processes. As a result, they form
the basis for developing customised intervention strategies tailored to various types of individuals and residents in different sectors, paving the way for effective and sustainable energy transition initiatives.
Palabras clave
Descripción
Materias
Matemáticas
Ciencia de los ordenadores
Sistemas de información, diseño y componentes
Ciencias Tecnológicas
Tecnología de los ordenadores
Dispositivos de transmisión de datos
Ciencia de los ordenadores
Sistemas de información, diseño y componentes
Ciencias Tecnológicas
Tecnología de los ordenadores
Dispositivos de transmisión de datos