A mathematical model has been developed that predicts the cell population dynamics and production of recombinant protein and infective extracellular virus progeny by insect cells after infection with baculovirus in batch suspension culture. Infection in the model is based on the rate of virus attachment to suspended insect cells under culture conditions. The model links the events following infection with the sequence of gene expression in the baculovirus replicative cycle. Substrate depletion is used to account for the decrease in product yield observed when infecting at high cell densities. Model parameters were determined in shaker flasks for two media : serum-supplemented IPL-41 medium and serum-free Sf900II medium. There was good aggreement between model predictions and the results from an independent series of experiments performed to validate the model. The model predicted : (1) the optimal time of infection at high multiplicity of infection; (2) the timing and magnitude of recombinant protein production in a 2-L bioreactor; and (3) the timing and magnitude of recombinant protein production at multiplicities of infection from 0.01 to 100 plaque-forming units per cell. Through its ability to predict optimal infection strategies in batch suspension culture, the model has use in the design and optimization of large-scale systems for the production of recombinant products using the baculovirus expression vector system.