Understanding Occupant Behaviors in Dynamic Environments using OBiDE framework

Abstract

Occupants' movements and presence are fundamental and the prerequisites for any type of occupant behaviors' understanding which tells whether a building location is occupied, the number of occupants or an occupant with a specific profile in a certain location. Numerous studies have been conducted over the past few decades to model occupant behaviors stochastically for an improved understanding of their activities for different facility management applications. Despite many research efforts to model dynamic behaviors of building occupants, the understanding of their behaviors by incorporating the contextual information linked to the evolution of the building environment is still not adequately explored. The contextual information linked to locations in dynamic environments changes often over time in terms of position, size, properties and relationships with the environment. This changing building environment affects the occupants' movements and presence inside the facility which ultimately degrades the process of inferring their accurate activities based on the location context. Henceforth, the evolving building information is required to be mapped with occupant movements for an improved understanding of their changing behaviors. To fill this research gap, a framework named 'Occupant Behaviors in Dynamic Environments' (OBiDE) is designed for providing a 'blueprint map' to integrate existing DNAS (Drivers, Needs, Systems, Actions) ontology (i.e. a scheme to model occupant behaviors) with our semantic trajectory enrichment model to better understand the occupant behaviors by tracking the dynamicity of building locations. The proposed framework extends the usability of DNAS by providing a centralized knowledge base that holds the movements of occupants with relevant historicized contextual information of the building environment to study occupant behaviors for different facility management applications.

Modeling occupant behavior in buildings

“…2) Shallow features can be recognized well with ML but a difficulty in identifying context-aware activities of occupant behavior (e.g., cooking a meal) or extracting other dimensions of occupant behavior [20,[277][278][279].…”

“…3) In traditional approaches, extensive training data and labeled annotations are mandatory for supervised learning, but in real-world applications, most of the data remain unlabeled (unsupervised). Due to this, typical models are unadaptable to a diverse range of context-aware occupant actions and model configurations [20,45,[275][276][277][278][279].…”

Thermal comfort practices for precast concrete building construction projects: towards BIM and IOT integration

“…heat stress with moderate capital and operational costs) without hindering normal performance activity of PC building construction projects. Arslan et al (2019) pointed out that the presence of IoT–BIMs in the management of PC building construction can improve compliance with environmental control and better optimise physical comfort, with designers and manufacturers considering both resource-optimum and system-optimum location choice behaviours and other requirements:Design and risk assessment of the integrated IoT-BIMs – occupancy schedule modelling system in construction projects;The engagement of occupant behaviours in building environments in terms of heat production and through energy consumption as well as location planning processes;Drivers-Needs-Actions-Systems (DNAS) ontology–building energy management planning that divides behaviour into movements, occupancy and body postures;Trade-off minimisation to enable sustainable building environments (e.g. minimal thermal injury or constructional damage effect) of sections of the construction project using PC while occupants' productivity, environmental quality in the building and functionality of building spaces for other sections are being considered.The employment of occupancy localisation and the validation of skill performance in filtering building layouts for important information, building scale (i.e.…”

“…POE findings provide benchmark criteria for comparing one facility's quality of finish, services and performance against another's and offer guidance to improve future developments. However, benchmarking facility performance via POE is problematic due to industry reservations that any value accrued is largely beneficial to industry competitors vis-a-vis the developer commissioning the evaluation (Arslan et al , 2019). Edirisinghe and Woo (2020) stated that cloud-based post-occupancy evaluation technology (e.g.…”

Impact of Window-to-Wall Ratio on Heating Demand and Thermal Comfort When Considering a Variety of Occupant Behavior Profiles

“…), occupant presence or movement, or operation conditions such as window opening, lighting, appliances use, temperature setpoints, heating demand, and more. Among the different models proposed in literature (Yan et al, 2017), some treated individual aspects like occupants' movements and presence (Arslan et al, 2019) DHW, electricity, window opening (Rouleau and Gosselin, 2020) or a combination of some of them (Rouleau et al, 2019b). Rouleau et al (2019a) studied robustness of energy consumption and comfort in a high-performance residential building with respect to occupant behavior.…”