智能工作流模型及驱动研究

Research on intelligent workflow model and drive

基于工作流参考模型提出了智能工作流模型,并根据该模型给出了一种通用、灵活且适应性强的工作流流程控制方法——弧线控制法.该方法仅需要几种简单的元素(条件集、流向)就可实现不同流程类型的流转,并且可以很容易地被引擎解释及控制;研究了工作流智能处理的模型、原理、方法及实现,提出了任务处理引擎(仲裁)及任务处理信息库,通过仲裁的处理能力及信息库信息的完备性和可扩充性最终实现了工作流的智能性;通过全文检索等技术对自然语言描述的任务选择最佳执行方法并进行自动处理,增强了工作流智能水平.

An intelligent workflow model based on Workflow Reference Model as well as a general, agile and adaptive workflow control method-Arc Line Control Method are proposed. This method requires only a few simple elements, such as Condition Sets and Flow Direction, but can solve the transition problem of different process patterns and also can be easily explained and con trolled by engine. Additionally, research on model, principle, method and implementation of Workflow intelligent Handle is carried out and Task Processing Engine （Arbitration）as well as Task Processing Information Library （TPIL） is put forward. Workflow Intelligence is implemented by task processing ability of arbitration and the integrality and expandability of TPIL. Selecting optimal execution method and processing automatic handling of tasks described with natural language are realized through global search, which enhances and reinforces the workflow intelligence.