A dual-paradigm approach to deriving optimal principal trade proportion thresholds for construction procurement classification

South Korea recently reformed its construction production system to integrate general and specialty contractors, aiming to remove partition barriers and enhance competitiveness. However, whether the reform achieves its intended outcomes remains insufficiently understood, while procurement decisions continue to rely on monetary ceilings or administrative discretion rather than the actual technical composition of projects. This study addresses this gap by deriving an empirically grounded threshold for the principal trade proportion, defined as the share of a project accounted for by its dominant trade. Using 1,494 general construction bid records following the opening of mutual market entry (2023–2025), the study validates the threshold through a dual-paradigm convergence framework comprising two independent unsupervised classification methods: a non-parametric, density-based approach and a parametric, probabilistic approach. The approaches yielded thresholds of 72.46% and 71.70%, respectively, converging within 0.76 percentage points. This convergence indicates that the boundary reflects an intrinsic structural property of the data rather than a methodological artifact. Based on these findings, the study proposes a threshold-based procurement governance model that flags projects exceeding the threshold for reclassification from general to specialty procurement, providing ordering bodies with an objective criterion for project classification and expanding prime-contracting opportunities for specialty contractors.