A fusion of data analysis and non-monotonic reasoning to restore missed RFID readings

Radio frequency identification (RFID) is a wireless technology which can efficiently track various items within certain proximity. It has the potential to become a great asset across many applications such as a tracking inventory within a warehouse and the ability to track medical utensils within a hospital environment. Unfortunately, there are several problems that hinder the wide scale adoption of RFID technology including the serious threat of missed readings. Current state-of-the-art methodologies which attempt to solve the problem of false negatives can still not effectively restore the data set completely. In this paper, we propose an architecture that utilises a fusion of both intelligent data analysis of the observational records and a non-monotonic reasoning engine designed to determine the most likely values to restore. We then perform an analysis upon our methodology in which we discuss the adoption of our application.