Fuzzy Ambient Intelligence for Intelligent Office Environments Saifullizam Puteh, Caroline Langensiepen and Ahmad Lotfi School of Science and Technology Nottingham Trent University, Nottingham, United Kingdom Email: {saifullizam.puteh, caroline.langensiepen, ahmad.lotfi}@ntu.ac.uk Abstract—In this paper we present an ambient intelligence system for modelling and control of power use within an office environment. We define a multi-scale model of the office worker, consisting of a coarse grained office user profile, together with a fine grained characteristics model to summarise their behaviours. Sensor data gathered from individual offices are used to create these models. Collected data are fuzzified to produce meaningful and understandable descriptions of the worker’s daily activities. These models of the individuals form the basis of a Dynamic Power Usage Scheme (DPUS) to control individual offices. A trial version of the system has been implemented using individual academic staff offices in a university campus. Experiments have shown that even a simplistic user profile can lead to significant power savings from appropriate control of computers in an office environment. Index Terms—Ambient Intelligence, Fuzzy Control, Fuzzy Systems, Intelligent Environments, Energy Saving, Energy Man- agement I. I NTRODUCTION In modern office environments, personal computers (PC), lighting systems and heating/cooling system are the main energy consumers. Many companies would like to reduce their energy usage for two reasons. The first is that they worry about the environment and want to reduce the impact they have on it. The second reason, and most likely the one that companies care about most, is cost. For example, PCs waste a lot of energy due to being left on for long periods of time when not in use. Even though they have power management modes to reduce their energy consumptions when they are not in use, these are not always used. Some workplaces incorporate reactive systems such as passive infra-red (PIR) activated lighting, but these can be activated/deactivated inappropriately. Heating systems often work on the assumption of a 9:00am to 5:00pm presence, five days a week, whereas an individual office worker may have a different schedule, including long periods out of the office. The aim of the our research is to incorporate a Dynamic Power Usage Scheme (DPUS) in an office environment to control PCs, lighting and temperature based on individual office users. To achieve this, it is important to understand each user’s profile initially before adapting the environment to the users requirements. For example, instead of waiting half an hour for a timer to turn off the PC, the system could detect that the user is no longer in the room and the computer is not performing any work. This would cause the DPUS to then put the PC into an energy saving mode. The main focus of this paper is to create a fuzzy character- istics matrix to summarise the activities of an office user based on data collected from a sensor network. The profile matrix is presented as fuzzy values which indicate the likelihood of the sequence of activities in future. This paper is organised as follows: in Section II of this paper a short summary of the related work regarding Ambient Intelligence (AmI) in office environments is presented. In Section III proposed system architecture is explained followed by the pre-processing data stage explained in Section IV. Section VI explain the proposed technique for generating fuzzy user profiles. Details of our experiments are presented in Section VII. Section VIII draws some conclusions and discusses the direction in which the research needs to progress. II. RELATED WORK Jori Reijula et-al. [1] defined the concept of an intelligent work environment and evaluated its meaning from a user’s perspective. They take a multi-professional approach to fore- see the possibilities and challenges awaiting future users of intelligent working environments, including the importance of ergonomics when designing intelligent work environments of the future. The “intelligent and personalized-energy conservation sys- tem (iPower)”, developed by Ye et-al. [2] is a good example of how a smart environment can help to reduce power consump- tion in an office. Sensor nodes placed around the environment are able to monitor different aspects of the office working environment. When these different sensor measurements are combined, a picture of the environment can be processed which allows for sensible decisions to be made on how to control that environment. Using wireless sensors is appropriate as it allows them to be placed in good locations very easily. The concept of smart furniture is also introduced by the example of using a pressure plate on a chair to indicate if someone is sitting on it. This could be used to determine if a user is at their desk, and so could be using their PC. A sensor-based modelling and prediction of user behaviour in intelligent buildings is proposed in [3]. The proposed system connects behavioural patterns to a building energy and comfort management system, although most of the work is based on the simulation tools. Some researchers have investigated the comfort of the worker in work place. Haynes [5] points out that while there is sufficient evidence to support claims that