The Chinese hamster ovary (CHO) cell line producing interferon-γ (IFN-γ) exhibits a 2-fold increase in specific productivity when grown at 32°C compared to 37°C. Low temperature also causes growth arrest, meaning that the cell density is significantly lower at 32°C, nutrients are consumed at a slower rate and the batch culture can be run for a longer period of time prior to the onset of cell death. At the end of the batch, product concentration is doubled at the low temperature. However, the batch time is nearly doubled as well, and this causes volumetric productivity to only marginally improve by using low temperature. One approach to alleviate the problem of slow growth at low temperature is to utilize a biphasic process, wherein cells are cultured at 37°C for a period of time in order to obtain reasonably high cell density and then the temperature is shifted to 32°C to achieve high specific productivity. Using this approach, it is hypothesized that IFN-γ volumetric productivity would be maximized. We developed and validated a model for predicting the optimal point in time at which to shift the culture temperature from 37°C to 32°C. It was found that by shifting the temperature after 3 days of growth, the IFN-γ volumetric productivity is increased by 40% compared to growth and production at 32°C and by 90% compared to 37°C, without any decrease in total production relative to culturing at 32°C alone. The modeling framework presented here is applicable for optimizing controlled proliferation processes in general. © 2003 Wiley Periodicals, Inc.