An increasing barrier to productivity in knowledge-intensive work environments is interruptions. Interruptions stop the current job and can induce forgetting in the worker. The induced forgetting can cause re-work; to complete the interrupted job, additional effort and time is required to return to the same level of job-specific knowledge the worker had attained prior to the interruption. This research employs primary observational and process data gathered from a hospital radiology department as inputs into a discrete-event simulation model to estimate the effect of interruptions, forgetting, and re-work. To help mitigate the effects of interruption-induced re-work, we introduce and test the operational policy of sequestering, where some service resources are protected from interruptions. We find that sequestering can improve the overall productivity and cost performance of the system under certain circumstances. We conclude that research examining knowledge-intensive operations should explicitly consider interruptions and the forgetting rate of the system's human workers or models will overestimate the system's productivity and underestimate its costs.