The building of excavations is an extremely dangerous job that incorporates a variety of different variables. It is possible to significantly lower the likelihood of an accident occurring by first accurately identifying high-risk variables and then taking appropriate preventative steps. Single-valued neutrosophic verbal sets (SVNVS) can effectively represent qualitative and vague information when used in the identification process for high-risk variables of excavation systems. In addition, the identification of highrisk elements associated with an excavation system is a multi-criteria decision-making (MCDM) issue. This issue may be resolved by using the multi-attribute border approximation area comparison (MABAC) technique. The MABAC method operates on the presumption that criteria are compensating. However, the identification process for high-risk variables of excavation systems may include characteristics that are not compensatory. Under conditions of single-valued neutrosophic sets, a MABAC approach is developed. The weights of the criterion are calculated using this approach, which uses the mean-squared deviation weight method. In addition to that, an illustrated example is carried out to demonstrate the process that is involved in the MABAC approach.