Designing context-aware systems: A method for understanding and analysing context in practice

Abstract

Context-aware systems are systems that have the ability to sense and adapt to the environment. To operate in large-scale multi-stakeholder environments, systems often require context awareness. The context elements that systems in such environments should take into account are becoming ever more complex and go beyond elements like geographic location. In addition, these environments are themselves so complex that it is hard to determine what parts of them belong to the relevant context of a context-aware system. However, insight into what belongs to this context is needed to establish what the design of a context-aware system should be to meet its goal. The ambiguity of what belongs to context in these complex organizational environments causes the design process to become either inefficient or less effective. In this paper, we provide a method to identify what elements of the environment are relevant context and to then base the design on this insight. The proposed method consists of three steps: 1) getting insight into context, 2) determining what components are needed to sense and adapt to context, and 3) determining the rules for how the system should adapt in different situations. To reduce ambiguity by organizations, the method requires a more specified definition of context than the ones in current literature, which we also provide in this paper. In addition to reducing ambiguity, the highly structured way in which the components and rules are derived from insight into context provides a way to further deal with the high complexity of the context. The method was applied for the development of a context-aware system for business-to-government (B2G) information sharing in the container shipping domain. Information sharing in this domain is highly complex, as legislation, many stakeholders, and a mix of cooperation and competition result in a highly complex environment. The development of this B2G information sharing system thus provides an example of how the method can be used to develop a context-aware system in a highly complex environment.