To solve the low degree of generalization and serialization problem of aviation support equipment, a portfolio scheduling method based on importance assessment and the improved multi-population genetic algorithm (IMPGA) is proposed. This method combines the expert knowledge system and swarm intelligence system, exerting their respective advantages. Firstly, the portfolio importance assessment index of aviation support equipment is established. Then the index weight model based on entropy weight and grey cluster is constructed, meanwhile the comprehensive assessment value is regarded as priority sorting of portfolio to realize preliminary screening of portfolio scheduling. Finally, highly portfolio importance assessment and controllable cost scheme selection are carried out in many times of iteration based on IMPGA, with the ratio of portfolio importance comprehensive assessment value to optimization cost as its fitness. The portfolio scheduling example validates the effectiveness of the proposed method.