Author(s)

Zhilin Zhang, Shuang Han

Corresponding Author

Zhilin Zhang

Abstract

This paper proposes a strategy-oriented AI framework for dynamic pricing and vacancy early warning in small- and medium-sized industrial parks. Centered on causal elasticity estimation, multi-source data fusion, constrained optimization, and explainable risk stratification, the framework aligns revenue maximization with occupancy stability and policy compliance. A layered architecture—data, features, forecasting, pricing optimization, and vacancy risk management— supports continuous, safe online learning and human-in-the-loop governance. A calibrated case analysis indicates that AI-driven approaches can increase revenue, reduce vacancy, and stabilize price volatility relative to rule-based baselines, offering a pragmatic pathway for operators to transition from reactive adjustments to proactive portfolio steering.

Keywords

Artificial Intelligence; Dynamic Pricing; Vacancy Early Warning; Industrial Parks; SMEs; Demand Forecasting; Price Elasticity; Causal Inference; Reinforcement Learning; Real Estate Strategy