Mitigate Gender Bias in Construction: Fusion of Deep Reinforcement Learning-Based Contract Theory and Blockchain

Abstract

With the remarkable progress in teleoperation, physical fitness-based gender bias has become negligible within the construction sector. Nonetheless, the labor market remains male-dominated, posing tremendous unfairness toward females. In light of this, we developed a two-phase recruitment framework that utilizes blockchain, zero-knowledge proofs (ZKPs), deep reinforcement learning (DRL), and contract theory, aiming to enhance fairness, transparency, and automation. First, we devised a resume screening approach independent of gender to ensure fairness and alleviate gender bias in candidate assessment, by leveraging blockchain and ZKPs. In the second phase, we introduce a recruitment process that combines blockchain and DRL-based contract theory. This integration successfully mitigates gender bias that may arise from the self-disclosure property of contract theory. To evaluate the effectiveness of our proposed approach, we conducted comprehensive simulations from various dimensions. The results demonstrated the robustness and superiority of our method.