The traditional demand unconstraining estimation methods in airline industry only address the spill and recapture problem of customer demand in parallel nonstop flights, and they fail to consider the substitution effects in airline network between direct and connecting flights. Based on the ranking list of customer preference, the set of customer types in airline network was defined. Meanwhile, a network nonparametric discrete choice model considering strategic customer behavior was developed. In light of the incompleteness of the historical booking data in network environment, from the perspective of online and offline trading platforms, the complete data log-likelihood functions under the conditions of uncensored and censored demand were established respectively. The EM algorithm was applied to jointly estimate the customer arrival rate and the probability mass function. After that, the unconstraining estimation calculation methods of primary demand, recapture demand as well as spill demand of customers in airline network were proposed. The feasibility and effectiveness of the proposed methods were verified by numerical simulations. Compared with the existing methods, they can accurately reflect the impact of substitution effects in network between products on customer choice behavior, and thereby more effectively avoid overestimating the primary demand of historical customers.