ABSTRACT
This paper addresses context aware information supply for information processes in the construction industry. The planning and creating of structures and buildings is based on collaborative construction information processes whose information supply is characterized by a decentralized exchange of heterogeneous application models. Throughout a construction project numerous architecture, engineering and management information models are created and used because of the inherent complexity of the building product and processes as well as the corresponding high number of involved domains. The various specialized applications e.g. CAD, FEM, cost or scheduling software and standards used lead to a weak software integration and inconsistent data handling. Therefore a central challenge in construction projects is still the management of the inhomogeneous distributed information environment where domain models data need to be adequately transformed, exchanged and managed horizontally (between client and contractor and among various discipline specific representations), longitudinally (in their temporal development along the project phases), and vertically (on different levels of abstraction).
To bundle various application models and indicate their dependencies the Multimodel as a kind of an information space was developed. Increased requirements for specialized work processes as well as the growing complexity of construction projects lead to an increase in the amount, scope and complexity of the exchanged information spaces. The corresponding Multimodels quickly become very large and unwieldy. This leads to an inhibitory effect on the construction information processes. The transfer of Multimodels and the orientation in the information spaces will be more complicated. In addition, this effect prevents a mobile use of Multimodel based information spaces. However various construction information processes for planning and creating of structures and buildings usually doesn’t need the entire information available in construction projects. When considering the situational information requirements of construction information processes, a context dependence is revealed that determines the quantity and quality as well as the cutouts and linking depth of the required information spaces depending on various aspects of the processing context. The information needs depends on the context. For an efficient information supply of construction information processes by matching information spaces therefore the consideration of the processing situation is necessary. That is why the design of the required Multimodels is determined by various aspects of the process situation. With knowledge of these aspects, context dependent information needs can be anticipated and targeted information supply of construction information processes can be realized.
A prerequisite for such a context aware information logistics is a methodology that enables formal depict dependencies between contextual information and information space elements and evaluate them automatically. For this purpose, in this work an approach is presented to formalize the context dependencies of the information needs of construction information processes. The influence of information logistically relevant context aspects to the design of information needs, the context relations, can be described directly in Multimodel templates. For this a special rule language, called ContextScript was developed that can be annotated instead of static attribute values into the template. In this way, based on Multimodel information spaces, context adaptive Multimodel templates can be defined, which can persisted along with an associated reference process. By evaluating these templates at time of use, situative information needs can be anticipated as MultiModel Template (MMT) and an adequate context oriented Multimodel can be generated. Thus the presented approach allows the realization of a context oriented information supply that allows to anticipate context based information needs and to generate a corresponding situationally information space.
