Scalable Enterprise Systems: Exploratory Research on Reconfigurable and Scalable Enterprise Planning Systems

这项探索性研究拨款用于开发正式方法，使信息系统设计人员和系统经理能够为电子商务世界中的企业资源规划(ERP)和制造执行系统(MES)设计和实施软件。形式主义将对企业系统采取控制论的观点，并将建立在以下原则之上：管理企业运营的业务规则应该仅软连接到支持ERP和MES的信息系统的设计。为企业运营的可重构、可扩展的软件系统提供了必要的理论基础。将开发方法，以工作流系统的形式获取业务流程和规则，并将其转换为有限自动机。将开发基于有限自动机和离散事件系统理论的形式化方法来合成控制律，如果将其施加到企业运营中，将确保符合指定的业务规则。将开发出理论结果，如果将其包括在综合过程中，将为企业软件系统提供诸如容错和从故障中自我恢复等重要特性。将开发分析方法来研究相互冲突的业务规则、业务规则和要求中的冗余以及多功能代理的分配。在研究结束时，将开发一个原型系统，包括前端，使用传统的过程建模工具，如DFD/结构化分析和UML来演示形式化。该原型将在具有电子商务能力的现有企业上进行Beta测试。如果成功，本研究的结果可用于构建通用软件环境，该环境包括综合、分析和优化，同时提供实施蓝图。软件将允许动态信息系统设计，并将为管理层提供控制法，以对企业中正在进行的物理流和信息流的操作施加一套不断变化的业务规则。在前端，将为IT工程师或企业经理提供更传统的建模工具，用于结构化分析、业务流程建模和需求规范。向有限自动机形式主义的转换将是自动的。在这个软件的核心将有一个计算引擎，它将合成控制律并分析它们以获得所需的特性。可以建立更多的模块来翻译这些控制法，并将它们提供给ERP、MES和支持数据库。

This exploratory research grant provides funding for the development of formal methods that allow information system designers and system managers to design and implement software for Enterprise Resource Planning (ERP) and Manufacturing Execution Systems (MES) in e-commerce world. The formalism will take a control-theoretic view of enterprise system and will be built on the principle that the business rules, which govern the operation of the enterprise, should only be soft-wired to the design of the information system, which supports ERP and MES. Furthermore, the formalism will provide the necessary theoretical foundation for re-configurable and scalable software system for enterprise operation. Methods will be developed to take the business process and rules in the form of workflow system and translate them into finite automata. Formal methods will be developed, based on the theory of finite automata and discrete event systems, to synthesize control laws, which if imposed on the enterprise operation will guarantee conformance to the specified business rules. Theoretical results will be developed, which if included in the synthesis procedure will provide the enterprise software system with such important properties as fault tolerance and self-recovery from failures. Analytical methods will be developed to study conflicting business rules, redundancy in the business rules and requirements, and allocation of multi functional agents. At the conclusion of this research, a prototype system including the front end featuring the use of traditional process modeling tools such DFD/structured analysis and UML will be developed to demonstrate the formalism. The prototype will be beta sited on an existing enterprise with e-commerce capabilities.If successful, the results of this research can be used to build a generic software environment, which encompasses synthesis, analysis and optimization while providing a blue print for implementation. Software will allow for a dynamic information system design and will provide the management with the control laws for imposing changing set of business rules over the operation of the physical flows and information flows going on in the enterprise. At the front end, the IT engineer or enterprise manager will be provided with more traditional modeling tools for structured analysis and business process modeling and requirements specification. The translation to finite automata formalism will be automatic. At the core of this software there will be a computational engine, which will synthesize control laws and analyze them for desirable properties. Additional modules can be built to translate these control laws and feed them to ERP, MES and the supporting database.