Office environments play an important role in fostering a productive and healthy work atmosphere. Based on the increasing use of smart technologies in office environments, data-driven approaches can be implemented for cleaning operations. Smart cleaning strategies can adapt to changing environmental conditions and usage patterns. This helps to make sure that the office environments are clean and hygiene by performing cleaning operations based on visitors count and weather conditions. The objective of this thesis is to design a predictive model that can support smart cleaning strategies by analyzing historical data from wi-fi networks, sensors, weather data, and other relevant sources. Using design science research methodology, this thesis aims to enhance cleaning operations decisions such as when and where to clean, when to refill resources, and dispose the garbage. By applying advanced data analytics on historical data specifically in the context of office cleaning, the findings of this research provides new knowledge into the field of facility management . The research findings also offer practical advices for organizations looking to adopt data analytics models to support their cleaning operations. By offering a model to support office cleaning, this thesis provides guidance for cleaning operations in smart buildings.