Risk detection and prediction from indoor tracking data

Abstract

Technologies such as RFID and Bluetooth have received considerable attention for tracking indoor moving objects. In a time-critical indoor tracking scenario such as airport baggage handling, a bag has to move through a sequence of locations until it is loaded into the aircraft. Inefficiency or inaccuracy at any step can make the bag risky, i.e., the bag may be delayed at the airport or sent to a wrong airport. In this paper, we discuss a risk detection and a risk prediction method for such kinds of indoor moving objects. We propose a data mining methodology for detecting risk factors from RFID baggage tracking data. The factors should identify potential issues in the baggage management. The paper presents the essential steps for pre-processing the unprocessed raw tracking data and discusses how to deal with the class imbalance problem present in the data set. Next, we propose an online risk prediction system for time constrained indoor moving objects, e.g., baggage in an airport. The target is to predict the risk of an object in real-time during its operation so that it can be saved before being mishandled. We build a probabilistic flow graph that captures object flow and transition times using least duration probability histograms, which in turn is used to obtain a risk score of an online object in risk prediction.