Good set-points make good neighbors: user seating and temperature control in uberized workspaces

Abstract

"Uberized" office spaces are becoming more common. In such co-working office spaces, occupants pay as they use for desks or cubicles. While uberization increases the economic utility of an office space, it can potentially conflict with a key goal of a smart office: personalized user comfort. In particular, an uberized smart office should handle users with conflicting thermal comfort preferences. Existing works have addressed either user comfort in spaces with hard partitions and single actuators or HVAC control to save energy in open spaces by detecting or predicting occupancy. We consider the decision problem of assigning users based on their comfort preferences to desks in an open-plan shared space without partitions. The problem is computationally challenging due to the assignment decision and the non-linearities in thermodynamic constraints. We use a thermal model to assign users to desks. We evaluate our approach in a real-world shared office setting through simulations. We find that our approach identifies the minimum energy configuration among those that maximize user comfort.