No abstract

Justify why the probability of misloading an assembly is based upon data for dry transfer operations rather than wet transfer operations since the industry data is available for wet transfer (see CAL-WHS-MD-000003 and CAL-WHS-MD-000001). RESPONSEThe analysis performed in Commercial Spent Nuclear Fuel Waste Package Misload Analysis (BSC 2003) is based on generic human error probabilities developed in NUREG/CR-1278, Handbook of Human Reliability Analysis with Emphasis on Nuclear Power Plant ApplicationsFinal Report (Swain and Guttmann 1983). These error probabilities are not specific to dry transfer operations. Rather, they relate to generic human operator errors of selection and commission.However, Commercial Spent Nuclear Fuel Waste Package Misload Analysis (BSC 2003) includes a sensitivity calculation in Section 6.3.1 that employs operator selection error probabilities based on the actual wet transfer operational data developed in Waste Package Misload Probability (BSC 2001). The results of this sensitivity calculation show that for the operative sequences 3C and 10C (the introduction of the transportation, aging, and disposal (TAD) canister design obviates the other sequences considered in this analysis which involve incorrect waste package selection), the 3C mean probability is unchanged and the 10C mean probability is lower (1.83 × 10 −6 per waste package; BSC 2003, Table 11). Hence, the use of actual wet transfer data results in a lower combined probability for the operative sequences 3C and 10C. Therefore, the more limiting result based on generic human operator errors of selection and commission was conservatively used as the basis for the postclosure criticality screening analysis. COMMITMENTS TO NRCNone. DESCRIPTION OF PROPOSED LA CHANGENone. Provide assessment of the effect of wrong documentation on misload probability. One assumption (Bullet 5.8) is that the documentation for use in the waste package loading process is correct. The study scope of CAL-WHS-MD-000003 does not analyze the process for generating the loading document. RESPONSEThe assumption of correct documentation pertains to the base case analysis performed in Commercial Spent Nuclear Fuel Waste Package Misload Analysis (BSC 2003), which is based on generic human error probability data as discussed in the response to RAI 3.2.2.1.2.1-4-013.The sensitivity analysis performed in Section 6.3.1 of Commercial Spent Nuclear Fuel Waste Package Misload Analysis (BSC 2003) and described in the response to RAI 3.2.2.1.2.1-4-012 relaxes this assumption by using actual industry wet transfer operational data for human error probability input. These data implicitly include the effects of wrong documentation amongst all the potential sources of error which resulted in the observed loading violations. Hence, the sensitivity analysis described in the response to RAI 3.2.2.1.2.1-4-012 provides an assessment of the effect of wrong documentation and demonstrates that DOE's estimate of the reliability of waste package loading is conservative. COMMITMENTS T...

NRC ABSTRACTA central question for resolution of GSI-24 is whether or not PWRs that currently rely on a manual system for ECCS switchover to recirculation should be required to install an automatic system. Risk estimates are obtained by reevaluating the contributions to core damage frequencies (CDFs) associated with failures of manual and semiautomatic switchover at a representative PWR. This study considers each separate instruction of the corresponding emergency operating procedures (EOPs), the mechanism for each control, and the relationship of each control to its neighbors. Important contributions to CDF include human errors that result in completely coupled failure of both trains and failure to enter the required EOP.This detailed study finds that changeover to a semiautomatic system is not justified on the basis of cost-benefit analysis: going from a manual to a semiautomatic system reduces the CDF by 1.7~10-~ per reactor-year, but the probability that the net cost associated with the modification being less than $1,000 per person-rem is about 20% without license renewal. Scoping analyses, using optimistic assumptions, were performed for a changeover to a semiautomatic system with automatic actuation and to a fully automatic system; in these cases the probability of having a net cost being less than $l,OOO/person-rem is about 50% without license renewal and over 95% with license renewal.