With the approval of ISA/ANSI S84.01, more and more efforts are necessary to meet desired safety targets. Statistical data are used to calculate the overall system availability or SIL. Failure rates depend on many factors, including the function of the equipment in the system and the definition of failure, the process environment and the maintenance practices, and the type of equipment and its manufacturer. In the process industry, the operating conditions and environments can change dramatically for the same equipment, and the range of failure rates observed can be quite wide. Borrowing data from laboratory and generic data sources introduces an element of uncertainty. Using point values may result in misleading evaluation of safety integrity level (SIL) of a safety instrumented system (SIS). The impact of data uncertainty on the calculation of SIL is discussed in this paper, and procedures to deal with data uncertainty in determining SIL for an SIS is proposed.