Assessing service system vulnerability in U.S. counties: a Pareto-based framework for integrating socio-demographic, built environment, energy, and health dimensions

Communities across the United States experience different levels of vulnerability to disruptions caused by natural hazards. These differences arise not only from hazard exposure but also from the underlying conditions of the systems that sustain everyday community functioning. While existing vulnerability assessments have made important advances in identifying socially vulnerable populations, few frameworks incorporate energy service conditions and health dependence into a systemic vulnerability assessment. The combined influence of four interacting dimensions, including socio-demographic conditions, built environment stress, energy service constraints, and population health dependence, remains poorly understood at the national scale. This study investigates how vulnerability emerges through the accumulation of constraints across these dimensions using a Service System Vulnerability Framework (SSVF). Principal Component Analysis (PCA) is applied to construct four subindices, and spatial statistics are used to identify geographic clustering of vulnerability. To avoid masking severe weaknesses through compensatory aggregation, the subindices are combined using a non-dominated Pareto ranking, producing a structured vulnerability classification for 3108 U.S. counties. The results reveal strong spatial clustering, particularly in the Lower Mississippi Delta, Central Appalachia, South Texas, Southern California, and several major metropolitan corridors. In many of these regions, vulnerability is driven not by a single dimension but by reinforcing interactions among socio-demographic disadvantage, built environment pressures, energy service stress, and health burdens. The Pareto analysis further shows that counties in the highest vulnerability tiers remain highly ranked because weaknesses in one service system cannot be offset by relatively better conditions in others. These findings demonstrate that vulnerability is better understood as a systemic condition arising from interacting service systems rather than as a single social or economic deficit. Consequently, this study advances a more holistic and policy-relevant understanding of vulnerability that can support equitable planning, preparedness, and resource allocation across diverse hazards and contexts.