Cognitive Facilities Management: Definition and architecture

Abstract

Prevailing facility management (FM) discourses increasingly recognise the importance of facilities in improving the quality of our life and the productivity of our core business. Nevertheless, no matter how intelligent or smart they are advocated, facilities are, by and large, treated as ‘aliens’ to us, human beings. This problem can be reduced by enhancing cognition ability of FM. Cognition is the process of acquiring knowledge and understanding through perceptions, action, and learning. Cognitive facilities management can percept through the cognitive Internet of Things, learns in the manner of human cognition with the power of cognitive computing, and acts actively, adaptively, and efficiently via robotics or automated actuators to improve the quality of people’s life and productivity of core business. The architecture of Cognitive FM proposed in this study consists of eight layers, namely, object layer, sensing layer, data layer, communication layer, computing layer, application layer, actuation layer, and evaluation layer, sequentially from lower to upper. Through these layers, two information cycles take shape. The perception-action cycle between object and actuation layer enables targeted response, whilst evaluation-adapting cycle between evaluation and computing layer facilitates adaptive learning. With the architecture and technologies behind, the CFMS attains three smart properties, i.e., awareness, communicativeness, and autonomy. The contributions of this Cognitive FM architecture are twofold: (1) promotes a potential approach towards advanced intelligence in FM; and (2) provides a detailed architecture to develop practical, smart FM applications.