Study on “digital intelligence + refinement” management mode for architectural laboratories

Architectural laboratories play a central role in practical teaching and scientific research in architecture, urban planning, and related disciplines. However, their management suffers from challenges related to complex spatial utilization patterns, diversity of used professional equipment, and discipline-specific safety risks. Traditional management modes are characterized by fixed partitioning, manual registration, and regular maintenance, which lead to low space utilization, a high idle rate of large-scale instruments, lagging early safety warnings, and poor adaptation to interdisciplinary experimental needs. To solve these problems, this study developed an integrated “digital intelligence + refinement” management mode for architectural laboratories. It aimed to break through efficiency bottlenecks with digital technologies and resolve landing pain points with refined systems to achieve the collaborative optimization of space utilization, equipment utilization, and safety management and provide a replicable reference for the management upgrading of similar laboratories in universities.

This research adopted a combined approach of case comparison and empirical iteration. First, through case comparison, management practices of architectural laboratories in comprehensive universities, engineering universities, and professional art academies were analyzed to identify common problems and disciplinary characteristics. Second, by taking the School of Architecture at Soochow University as the empirical object, a phased iterative practice was adopted. In the first stage, a building information modeling (BIM)-based three-dimensional (3D) modeling platform was built to link equipment parameters, operational procedures, and spatial demand rules. In the second stage, an intelligent closed-loop safety system, including intelligent cabinets for hazardous chemicals, artificial intelligence (AI) visual monitoring, and access control linkage, was deployed to realize whole-process traceability and real-time risk identification. In the third stage, refined supporting mechanisms were established, such as flexible spatial division, hierarchical safety responsibility, equipment full-life-cycle management, and professional team training. Mode effectiveness was dynamically evaluated and optimized using indicators such as the space utilization rate, safety incident rate, and equipment sharing rate.

The empirical results showed that the proposed “digital intelligence + refinement” mode achieved significant application outcomes. BIM and Internet of Things technologies enabled dynamic spatial scheduling and real-time environmental monitoring. The waiting time for equipment in the model laboratory was reduced from more than 2 hours to 25 minutes, and the space could be quickly reorganized within 15 minutes to allow multi-scenario experimentation. In terms of safety management, the intelligent monitoring system accurately identified high-frequency, risky behaviors in architectural experiments with a recognition accuracy of 98% and realized graded early warning and closed-loop disposal. The standardized management of hazardous chemicals with low toxicity and high volatility eliminated illegal storage and misuse. For equipment management, a digital ledger documenting purchased and self-developed instruments was established to enable full-life-cycle tracing and priority-based maintenance, greatly reducing the failure rate of self-developed devices and improving the overall utilization rate of instruments. Meanwhile, the refined training and assessment system enhanced the comprehensive competence of technical staff in BIM operation, equipment calibration, and safety control, ensuring the sustainable implementation of digital systems.

The “digital intelligence + refinement” integrated management mode effectively solves the prominent issues in the management of architectural laboratories. It forms a 3D integrated system supported by digital technologies, guaranteed by refined systems, and adapted to discipline-specific scenarios, which considerably improves space utilization, reduces potential safety hazards, and optimizes resource allocation. The phased, iterative implementation path is especially suitable for architectural laboratories with complex functions and difficult reconstruction. In the future, the application of generative AI and digital twins can be further explored to achieve virtual–real integration, remote monitoring, and intelligent prediction, continuously improving the intelligence level and service capacity of architectural laboratory management.