This paper provides a two stage stochastic model for an international air cargo forwarding problem under uncertainty. The air forwarders have to make a decision about the number of containers to be booked for the hob and regions in advance before accurate customers information becomes available, which is called the first stage decision. In the second stage, when the accurate information is realized, the forwarders have to decide the number of extra containers to be required or the containers to be returned, as well as cargo loading plans for each scenario that might happen on the shipping day. Computational results show that the stochastic model can provide a cost-efficient, flexible and responsive container booking and cargo forwarding system.