Green Production Cost Audit Model for Textile Enterprises via IoT Multi-Source Data Fusion

The textile industry, as a pillar of the Chinese economy, faces persistent challenges related to high energy consumption, environmental pollution, and inadequate environmental cost accounting under conventional auditing approaches that primarily emphasize economic indicators. To address these issues, this study proposes a green production cost auditing model for textile enterprises based on IoT-enabled multi-source data fusion. The proposed framework integrates a three-tier IoT architecture (sensing– transmission–platform) to acquire energy, environmental, and process information through heterogeneous sensing and wireless communication networks, providing reliable data support for intelligent industrial monitoring. An enhanced Segmental-DTW algorithm is employed to resolve temporal asynchrony among multi-source data, while a temporal-LSTM network dynamically weights process-specific features. Hidden environmental costs, including carbon emissions and pollution risks, are quantified through Monte Carlo simulation within a multidimensional evaluation framework. Empirical results from a medium-sized textile enterprise in the Yangtze River Delta show that the green-audited unit cost (¥3.08/m) is 29.4% higher than that obtained using traditional methods (¥2.38/m), with hidden environmental costs accounting for 22.7%. Sensitivity-guided process optimization further reduces comprehensive energy consumption by 16.7%, wastewater discharge by 21.7%, and annual costs by more than ¥1,400,000. The proposed model provides a scalable solution for sustainable textile manufacturing while demonstrating the value of IoT sensing, heterogeneous data transmission, and intelligent information integration for advanced industrial systems, offering useful insights for electromagnetic sensing and wireless communication applications in smart production environments.