The construction of large underground caverns fundamentally differs from building and above ground civil infrastructure projects due to their complex geometries and variable geological conditions. These projects are complex and challenging because a large amount of data is generated from dispersed, independent, and heterogeneous sources. The underground construction industry often uses traditional project management techniques to manage complex interactions between these data sources that are hardly linked, and independent decisions are often made without considering all the relevant aspects. In this context, cavern construction exhibits uncertainties and risks due to unforeseen circumstances, an intricate design, and ineffective information management. Existing research has considered general BIM semantic models at the design stage; however, the digital transformation of cavern construction remains underdeveloped and fails to integrate digital construction throughout the project lifecycle. To address that gap, a novel BIM-based multi-model cavern information modeling framework is presented here to improve project management, construction, and delivery by integrating multiple interlinked data models and project performance data for large underground cavern construction. Data models of cavern construction processes are linked to propose an extension of the Industry Foundation Classes (IFC) schema based on the cavern-specific elements, relationships, and property set definitions. To illustrate the potential of the proposed framework, a theoretical application to the powerhouse cavern construction is presented. The results indicate that the framework has significant potential to improve construction efficiency and safety and establish a robust foundation for the digital transformation of underground cavern projects. The theoretical implementation on the Neelum–Jhelum powerhouse cavern showed that the framework enabled a 92 m cavern realignment to avoid fault zones, achieved a 12.4% reduction in rock bolt usage, and a 9.8% reduction in shotcrete volume. These quantitative improvements illustrate its potential to enhance safety, reduce material costs, and optimize construction efficiency compared to conventional workflows.

Efficient monitoring and recognition of excavator operational activities are critical for improving productivity, safety, and intelligent construction management. Traditional vision-based activity recognition systems remain sensitive to challenging construction site conditions, including occlusion, dust, variable illumination, and limited visibility. This paper presents an excavator activity recognition framework that, for the first time, applies Gramian Angular Field (GAF) encoding to multi-sensor excavator operational data, transforming them into spatial image representations and enabling deep convolutional neural networks (CNNs) to extract discriminative features. The proposed approach integrates synchronized excavator sensor signals including bucket positional coordinates, body orientation, fuel consumption, engine RPM, and joint angles to characterize operational behavior during four representative activities: digging, dumping, idle, and levelling. Unlike conventional sensor-based methods that directly process sequential time-series data, our GAF-based framework transforms operational signals into structured spatial representations that preserve temporal correlations while enabling effective feature learning through image-based deep learning architectures. Experimental results demonstrate that the proposed GAF-CNN-LSTM framework achieves 8.24 percentage points higher classification accuracy compared with LSTM networks, and 6.24 percentage points higher than 1D CNN baselines trained on the same sensor data. The method effectively captures discriminative operational signatures across multiple excavator activities in the collected dataset. This work bridges the limitations of both vision-based and traditional sensor-based approaches, providing a promising framework for excavator activity monitoring in construction and fleet management contexts, pending further validation under operational deployment conditions.