While airline schedule was disrupted, airlines needed not only to develop aircraft and crew recovery schedule to ensure flight operations, but also to recover passengers flow recovery in time to reduce economic loss caused by passenger delay or cancelation of itineraries. This paper set up a linear integer programming model for the passenger. The model’s variables were the passenger flows on alternative itineraries. In order to solve the model, after constructing a set of disturbed OD pairs, all available itineraries were generated by depth-first searching the flights network. Finally, the model was solved using simplex algorithm. Empirical results demonstrate the ability of the optimal model and algorithm in the paper to quickly and efficiently deal with the passenger recovery problem.