A Computational Decision Matrix for Sustainable Tourism: Machine Learning Archetypes and Digital Leapfrogging
The post-COVID-19 tourism recovery exposes a structural divergence between economic resilience and environmental sustainability. Traditional tourism planning metrics consistently fail to diagnose how macroeconomic growth dynamics decouple from environmental pressures, leaving policymakers without empirical tools to identify structural vulnerabilities or prevent carbon-intensive recoupling during post-crisis transitions. This study integrates macroeconomic, environmental, and digital data across a global panel to map actionable pathways for sustainable tourism transitions. Employing a multi-stage methodology, the analysis first utilizes K-Means clustering (n = 80) to isolate the structural fixed effects of baseline destination archetypes driving a K-shaped recovery. Second, using a synchronized environmental panel (n = 41), a Decoupling Index evaluates eco-efficiency elasticity to test the alignment between tourism value recovery and aviation-induced CO2 emissions. Third, regression analysis of an elite digital cohort (n = 18) measures dynamic exogenous catalysts, revealing that digital attractiveness, proxied by the global digital nomad market share, is a significantly stronger accelerator of recovery (β = 55.59, p = 0.019) than traditional physical air connectivity (β = −46.48, p = 0.036). Synthesizing these insights, a 2 × 2 Strategic Decision Matrix (n = 41) classifies destinations into Sustainable Leaders, Mass-Market Traps, Value Pivoters, and Vulnerable Laggards. The empirical results demonstrate that pre-pandemic structures do not deterministically dictate recovery (p > 0.05, Partial η2 ≤ 0.077), yet rapid financial recovery often masks deep atmospheric vulnerabilities, with specific absolute decoupling leaders achieving exceptional value expansion alongside strict carbon contraction (e.g., Saudi Arabia, DE = −0.35; El Salvador, DE = −0.26). This framework provides a data-driven roadmap for policymakers to utilize “soft” digital infrastructure to transition from carbon-intensive, volume-dependent models toward value-optimized, low-emission ecosystems.