A BIM-DFMA integrated framework for data-driven pre-assembly analysis of large-span steel truss structures

For complex steel structures, the substitution of physical trial assembly with virtual assembly has the potential to significantly reduce both manufacturing costs and time. However, current pre-assembly verification methods face challenges in accurately quantifying assembly deviations that arise from the accumulation of manufacturing tolerances. To address this issue, the present study proposes an innovative framework that integrates Design for Manufacturing and Assembly (DFMA) principles to facilitate efficient data exchange. By synthesising Building Information Modelling (BIM) engines with DFMA theory, a Pre-Assembly Analysis System (PAS) and a data-driven conflict detection algorithm have been developed to elucidate the cumulative impact of manufacturing deviations on assembly accuracy. Validation through the use of a large-span cantilever steel truss structure demonstrated that the PAS effectively performs essential functions, such as geometric inspection and assembly conflict identification, thereby significantly enhancing assembly precision. This research offers an efficient digital solution for lean construction in steel structures.