基于设计结构矩阵的航天项目解耦方法研究以载人航天工程项目Z为例

航天项目具有任务组成复杂、内部迭代繁多的特点，易出现因任务耦合导致严重延期的问题。现行航天项目管理方法无法直观表示项目任务耦合关系，难以对耦合度较大的任务进行识别、优化。本文提出将任务反馈跨度和任务反馈的前序任务与反馈任务数量之比两个因素构建耦合度评价的函数，利用设计结构矩阵任务排序解耦原理，结合遗传算法求解耦合度最低的项目任务排序的方法。最后以载人航天工程项目Z作为例，证明该方法的有效性。

Research on Decoupling Method of Space Project Based on Design Structure MatrixTaking Manned Space Project Z as an Example

Space projects have the characteristics of complex mission composition and various internal iterations, which are prone to serious delay due to mission coupling. The current space project management methods can not directly express the coupling relationship of project tasks, and it is difficult to identify and optimize the tasks with large coupling degree. In this paper, the function of coupling degree evaluation is constructed by two factors: the span of task feedback and the ratio of the number of tasks in front of task feedback to the number of tasks in feedback. Based on the decoupling principle of task sequencing of design structure matrix, and combined with genetic algorithm, the method of task sequencing with the lowest decoupling degree is proposed. Finally, taking the manned space project Z as an example, the effectiveness of the method is proved.