Assessment and prediction of human error rate plays the vital role in the probabilistic safety and risk analysis. Unfortunately no generalized method for human error classification and quantification exist. Prevailing methods are industry specific and give only a rough estimate. Underground coal mining is inherently hazardous and estimation of human error rate for mining activities is mandatory for analysis and enhancement of safety. Present study embraces the cognitive reliability and error analysis method for assessing the human error rate in reference to the system context. The present study captured the embedded uncertainty in data classification, context assessment and error categorization using fuzzy relational mapping and rough set theory.