A Comparative Objective Weighting MCDM-GIS Approach to Renewable Hydrogen Production Site Selection: A Spanish Case Study

Abstract

Deciding on the optimal location for a hydrogen production plant is a complex task that involves multiple factors, including proximity to demand centers, availability of renewable energy resources or existing infrastructure. Identifying suitable sites, therefore, requires integrating diverse geospatial criteria, typically addressed through Geographic Information Systems (GIS) and multi-criteria decision-making (MCDM) frameworks. Existing analyses rely on subjective weighting, which introduces biases and hampers replicability.Objective weighting methods offer a data-driven alternative but have generally been applied to selecting among only a few regional alternatives.This study fills this gap by developing a high-resolution spatial MCDM framework to evaluate thousands of candidate locations across Spain. Increasing spatial resolution from aggregated units (e.g., provinces) to site-level alternatives reveals greater variability in the input indicators, including a wider dispersion of values that is typically masked by spatial averaging. These differences may influence the resulting weights and rankings. To examine this issue, we compare five objective weighting methods using mathematical validation and an empirical comparison with the spatial distribution of existing hydrogen projects.Our results show that all methods are robust and consistently identify the most suitable areas, aligning well with existing project locations. The main differences between methods emerge in intermediate rankings and in their stability to changes in input data, with some methods being more sensitive to constraint definitions, such as the maximum allowable distance to demand centers. These findings highlight that, although objective methods substantially reduce the reliance on expert judgment, they do not eliminate it.Methodological choices materially influence prioritization outcomes, and informed expert judgment remains essential to ensure that methodological choices reflect the objectives and constraints of the analysis.

Deciding on the optimal location for a hydrogen production plant is a complex task that involves multiple factors, including proximity to demand centers, availability of renewable energy resources or existing infrastructure. Identifying suitable sites, therefore, requires integrating diverse geospatial criteria, typically addressed through Geographic Information Systems (GIS) and multi-criteria decision-making (MCDM) frameworks. Existing analyses rely on subjective weighting, which introduces biases and hampers replicability.Objective weighting methods offer a data-driven alternative but have generally been applied to selecting among only a few regional alternatives.This study fills this gap by developing a high-resolution spatial MCDM framework to evaluate thousands of candidate locations across Spain. Increasing spatial resolution from aggregated units (e.g., provinces) to site-level alternatives reveals greater variability in the input indicators, including a wider dispersion of values that is typically masked by spatial averaging. These differences may influence the resulting weights and rankings. To examine this issue, we compare five objective weighting methods using mathematical validation and an empirical comparison with the spatial distribution of existing hydrogen projects.Our results show that all methods are robust and consistently identify the most suitable areas, aligning well with existing project locations. The main differences between methods emerge in intermediate rankings and in their stability to changes in input data, with some methods being more sensitive to constraint definitions, such as the maximum allowable distance to demand centers. These findings highlight that, although objective methods substantially reduce the reliance on expert judgment, they do not eliminate it.Methodological choices materially influence prioritization outcomes, and informed expert judgment remains essential to ensure that methodological choices reflect the objectives and constraints of the analysis.