An innovation-driven model and multi-dimensional evaluation of industrial transformation in resource-based regions: A case study of the Yellow River π-shaped Curve Area

Industrial transformation in resource-based regions is constrained by path dependence, ecological pressures, and insufficient collaborative innovation. Existing studies have paid insufficient attention to dynamic transition mechanism and multi-dimensional efficiency assessment in these regions. Using the Yellow River π-shaped Curve Area as a case study, this paper constructed a three-dimensional "factor-temporal-actor" analytical framework and integrated dissipative structure theory with a DEA-Malmquist-BCC hybrid model to examine dynamic and static efficiency as well as spatiotemporal differentiation from 2010 to 2022. Findings revealed: ①Innovation levels exhibited multi-dimensional and uneven gradient evolution across the region. Growing cities developed a technology-finance dual-driven pattern through resource reinvestment, whereas provincial capitals experienced a mismatch between policy support and technological innovation. ②Innovation synergy exhibited both temporal growth and spatial differentiation. The synergy index of provincial capitals and regeneration-oriented cities remained significantly above the regional average. Growing cities achieved annual increases of 5.8%~6.0%, whereas declining cities faced rising marginal costs in improving innovation coordination. ③Industrial transformation capacity displayed significant spatiotemporal differentiation. The energy-rich northern region benefited from advantages in technological industrialization, whereas the ecologically fragile southern region was constrained by path lock-in and insufficient economies of scale. Industrial transformation trajectories could therefore be categorized into fluctuating-improving, fluctuating-declining, and fluctuation-rebound types. ④Industrial transformation total factor productivity (TFP) declined by an average of 2.3% annually, mainly due to the simultaneous decline in technical and scale efficiency. Spatially, the region exhibited a gradient pattern characterized by stronger performance in the east and north. By integrating stage-specific interventions, cross-scale governance frameworks, and dynamic feedback mechanisms, this study proposes a sustainable pathway for resource-based regions to suppress entropy growth, delay path lock-in, and stimulate industrial transition. The findings provide theoretical support and policy implications for overcoming transformation traps and promoting innovation-oriented coordinated development in the Yellow River Basin.