Occupational health and safety applications require a comprehensive risk analysis to protect employees from accidents caused by various hazard sources in the workplace. Therefore, in the literature, there are various risk assessment methods utilizing various parameters. One of them is safety and critical effect analysis (SCEA) which is the method that calculates risk magnitude based on the highest number of parameters that are probability, severity, frequency, and detectability. However, the risk assessment phase includes uncertainties and inconsistencies caused by both evaluations of experts and the complex structure of the handled process. Therefore, we proposed a new risk assessment approach to cope with these uncertainties and inconsistencies by extending SCEA with the Neutrosophic sets. In this study, each parameter, probability, severity, frequency, and detectability, is defined by using truth membership function, falsity membership function, and indeterminacy membership function in Neutrosophic sets (NSs) to deal with incomplete, indeterminate, and inconsistent information existing in the expert preferences and complex structure of the handled process. It is the first time, the indeterminacy of NSs is defined based on truth and falsity definitions by using a rule-based system based on probability, severity, frequency, and detectability parameters. Furthermore, Mamdani fuzzy inference system used in the SCEA method is adapted to NSs. The proposed approach is applied to the metalworking and woodworking workshop of a university. The obtained results are compared with the results of the SCEA method. It is concluded that the proposed approach is better than SCEA method to consider inconsistencies in the risk evaluation. (C) 2021 Elsevier B.V. All rights reserved.