No matching vaccine is immediately available when a novel influenza strain breaks out. Several nonvaccine-related strategies must be employed to control an influenza epidemic, including antiviral treatment, patient isolation, and immigration detection. This paper presents the development and application of two regional dynamic models of influenza with Pontryagin’s Maximum Principle to determine the optimal control strategies for an epidemic and the corresponding minimum antiviral stockpiles. Antiviral treatment was found to be the most effective measure to control new influenza outbreaks. In the case of inadequate antiviral resources, the preferred approach was the centralized use of antiviral resources in the early stage of the epidemic. Immigration detection was the least cost-effective; however, when used in combination with the other measures, it may play a larger role. The reasonable mix of the three control measures could reduce the number of clinical cases substantially, to achieve the optimal control of new influenza.

Many studies suggest that air-transportation networks contribute a lot to the spatiotemporal dynamics of infectious diseases. The mobility of individuals over the networks has greatly speeded up the spreading processes of the epidemics and pushed the population in non-epidemic areas into the risk of infection. To figure out the underlying interactions between the air-transportation networks and the transmission of the epidemics, we have (i) analyzed the air-routes and the statistical data on the passenger throughput of the civil aviation of China and (ii) carried out a computer simulation based on the assumption that a novel influenza outbreaks in Southeast Asia. The results show that the topology of the air-transportations networks has a typical structure of heterogeneities. We also find that the epidemics will soon strike China after the initial outbreaks and rapidly spread throughout the whole networks without air-travel restrictions even the reproductive number () is small.