基于改进GERT的任务过程时间特性建模分析方法

随着系统任务需求的日渐密集，用户对系统的时间特性提出了更为严苛的要求。首先，利用图示评审技术（GERT）构建系统任务流程的随机网络模型，针对任务中由于资源共享引起的活动排队执行以及部分上下游活动之间存在重叠的情况，基于排队论并引入时间因子以修正任务中各活动的执行时间，并给出任务执行时间的求解步骤。经考虑这2种情况，最终任务平均执行时间增加了22.9%。其次，通过不确定性分析，找出了任务中的关键活动，为进一步任务优化提供方向和思路。最后，以舰载机着舰任务为例进行案例应用分析，验证所提建模分析方法的有效性和适用性。 

Improved GERT based time characteristic modeling and analysis method for task process

The characteristic of time is an important requirement for system tasks, especially in complex industrial systems. This paper adopts Graphical Evaluation and Review Technique (GERT) to build the stochastic network model of task flow. Considering the queued execution of activities caused by resource sharing and the overlapping of some upstream and downstream activities in tasks, the queuing theory and time factor are introduced to modify the execution time of each activity in the task, and the solution steps are given. The average execution time of the task increases by 22.9%; through uncertainty analysis, moreover, the key activities in the task are found out, which can provide direction and ideas for further task optimization. The landing task of carrier-based aircraft is taken as a case study to demonstrate the effectiveness and applicability of the proposed method.