Agent-Oriented Modeling for a Dynamic and Evolving World

近年来，软件和信息技术(IT)的一波又一波的进步改变了社会。我们的工作、娱乐和社交互动现在错综复杂地由IT系统进行调节。然而，新兴技术即将引发新一轮转型。低成本的传感器网络、物联网、位置感知和移动设备，以及先进的数据分析和商业智能，当结合在一起时，将带来一场新的数字革命--感知和解读。当与为执行预定义程序和交互而高度开发的现有IT基础设施相结合时，新兴的企业级IT架构将帮助组织更快地响应和适应其日益快速变化的环境。 本研究项目旨在开发一个建模框架，以帮助概念化、分析和设计不同软件和IT系统的互补应用程序，以在高度动态和快速发展的环境中实现业务和组织目标。 在信息系统和软件工程中，系统设计人员和架构师主要依赖适当的模型来构思、描述、分析和管理复杂系统。需求工程领域已经开发了一些模型和技术，它们有助于将对系统的“业务”级别的理解与其技术设计和实现联系起来。近年来，该领域已从早期专注于单个系统的详细规范扩展到涵盖整个企业范围，从业务战略到政策和计划，再到业务流程，再到技术系统。在这个项目中，在认识到一系列新的“感知和解释”技术的迅速采用的同时，我们将开发一个新的建模框架来支持新的数字革命，在这场革命中，软件和数据将在感知和解释方面提供数量级的进步，就像过去的信息技术进步带来了数量级的执行效率提升一样。 数字革命的这一新阶段也得益于软件体系结构的进步，这些体系结构提供了更大的灵活性和适应性，如云计算和面向服务的计算，以及具有自适应能力的计算。本研究将开发的框架将指导企业范围的体系结构的开发，这些体系结构将利用感知和解释技术来利用新的软件体系结构提供的适应性，克服传统IT系统中常见的僵化。该框架将帮助系统设计人员和企业架构重新考虑设计时和运行时之间的边界，以及人类和自动化系统之间的责任重新分配。新的革命还可能改变软件工程过程，因为一些知识密集型活动(包括需求工程活动)可能会从设计时迁移到运行时，以利用来自环境的近实时输入。 该项目将建立、调整和大幅扩展面向代理的建模框架，该框架是面向目标的需求工程的领先框架。它强调了在复杂的软件和信息技术应用的背景下分析半自治行为者之间的战略利益和社会关系的必要性。I*框架源于申请者的博士论文，是国际标准的基础。拟议的研究将借鉴从战略管理到软件架构再到社会生态学等不同学科的概念。将使用加拿大工业界的案例研究来得出框架要求并验证结果。

Successive waves of advances in software and information technologies (IT) have transformed society in recent years. Our work, play, and social interactions are now intricately mediated by IT systems. Yet, newly emerging technologies are about to trigger another round of transformation. Low-cost sensor networks, the Internet of Things, location-aware and mobile devices, and advanced data analytics and business intelligence, when used in combination, will bring about a new digital revolution in “sensing and interpreting”. When coupled with existing IT infrastructures which are highly developed for executing pre-defined procedures and interactions, the emerging enterprise-wide IT architectures will help organizations become much more responsive and adaptive to their increasingly rapidly changing environments. This research project aims to develop a modeling framework to help conceptualize, analyze, and design the complementary application of diverse software and IT systems to achieve business and organizational goals in highly dynamic and rapidly evolving environments. In information systems and software engineering, system designers and architects rely crucially on appropriate models to conceive, describe, analyze, and manage complex systems. The field of Requirements Engineering has developed models and techniques that help bridge a “business” level understanding of a system to its technical design and implementation. In recent years, the field has expanded from an earlier focus on detailed specification of individual systems to encompass an enterprise-wide scope, connecting from business strategy to policies and plans, to business processes, to technology systems. In this project, in recognizing the rapid adoption of a new crop of “sense-and-interpret” technologies, we will develop a new modeling framework to support the new digital revolution, one in which software and data will offer orders-of-magnitude advances in sensing and interpreting, just as past IT advances had led to orders-of-magnitude gains in executional efficiency. This new phase of the digital revolution is also enabled by advances in software architectures which offer greater flexibility and adaptability such as cloud and service-oriented computing, as well as those with self-adaptive capabilities. The framework to be developed in this research will guide the development of enterprise-wide architectures that will leverage sense-and-interpret technologies to exploit the adaptability offered by the new software architectures, overcoming the rigidities common in traditional IT systems. The framework will help system designers and enterprise architectures rethink the boundary between design-time and run-time, and the re-allocation of responsibilities among humans and automated systems. The new revolution could also transform software engineering processes, as some knowledge-intensive activities (including requirements engineering activities) may migrate from design-time to run-time to leverage near-real-time input from the environment. The project will build on, adapt and substantially extend the i* agent-oriented modeling framework, a leading framework for goal-oriented requirements engineering. It highlights the need to analyze strategic interests and social relationships among semi-autonomous actors in the context of complex software and IT applications. The i* framework originated from the applicant’s PhD thesis, and is the basis for an international standard. The proposed research will draw upon concepts from diverse disciplines ranging from strategic management to software architecture to socio-ecology. Case studies with Canadian industry will be used to derive framework requirements and to validate results.