A Trade-Off-Aware Scoring Framework for Evaluating Alternatives under Conflicting Criteria: A Synthetic Stress Test and an Illustrative Application
Аннотация
This study proposes the Trade-off-Aware Scoring with Conflicts (TASC) model to evaluate alternatives under multi-criteria environments characterized by internal contradictions. By integrating Euclidean and cosine distances, veto threshold screening, and an exponential penalty layer governed by a structural Trade-Off Severity Index (TSI) into a unified multiplicative architecture, TASC systematically embeds conflict awareness into the evaluation process. Data-driven penalty thresholds and enforcement coefficients are calibrated directly from empirical indicator correlations. The framework is evaluated using a dual complementary approach: a synthetic dataset of 1,000 alternatives generated via a synthetic simulation procedure provides internal computational verification under extreme boundary stress, while an empirical dataset of 25 sovereign economies provides illustrative external evidence demonstrating practical policy-relevant viability. Results demonstrate that TASC effectively penalizes structurally asymmetric trade-off profiles and isolates structurally coherent solutions from those with latent strategic contradictions. Sensitivity analyses indicate strong ranking stability ($$\:\rho\:\ge\:$$ 0.9798) under varying penalty intensities and localized objective weight fluctuations ($$\:\pm\:$$ 10%). TASC provides a transparent, interpretable, and tunable framework that, within the scope of the tested scenarios, shows promise for complex public policy, sustainability screening, and infrastructure planning contexts where absolute performance excellence must be balanced against systemic structural harmony.