The objective was to devise a simulation model that could economically compare two group-housing systems for sow operations, a conventional system (CONV) that uses partial-slatted concrete flooring and an alternative system (ALT) that uses straw-covered concrete. Further, the research aimed to determine the optimal parity in which to terminally cull sows. Data were collected on 121 sows between two experimental barns for 7 parities. These data were used to estimate a production function and a culling function. These functions, along with economic data, were used to create an economic simulation model. Production was best predicted by parity and lactation feed intake, and culling was best predicted by parity, weight, and gait score. Optimal terminal culling occurred after parity 6 in ALT and after parity 7 in CONV. Overall, ALT was more profitable than CONV.