Abstract
Multi-sensor plays an important role in monitoring the systems and equipment under consideration. The final integrated information obtained from these sensors is always associated with uncertainties and ambiguities, thus to eliminate these inaccuracies, various uncertainty theories, e.g., Dempster–Shafer theory (DST), have been proposed. In several models, it is optimistically assumed that the sensors always work properly, but in reality, they can fail like any other physical equipment. In the past decades, many researchers calculated the dynamic sensor reliability by comparing the information given by each sensor to those of other sensors. However, such methods are sometimes problematic because there are chances while the considered sensor is working properly, the others can fail. In order to address these problems, our proposed methods introduce a sensor risk factor in the sensor output to enhance the accuracy. In this paper, we present two pessimistic–optimistic models based on the interval-valued and fuzzy-valued DST to model this error factor. Finally, a numerical example is given to exemplify the application of the proposed models