Design of BOM and Data Detection Algorithm for Railway Vehicles

The BOM architecture is proposed in this paper from the production process of railroad vehicles, combined with the current situation of BOM in a company. The construction methods of MBOM and SBOM are elaborated in this paper. In order to quickly and accurately detect outliers in BOM part number data, an unsupervised learning algorithm based on isolated forest is proposed and compared with the box plot. The results show that the unsupervised learning algorithm based on isolated forests performs better for outlier detection. This study meets the rapid development needs of railway vehicles and provides a reference for other companies to solve BOM management problems.