HOME > About the azbil Group > Research & Development > Direction for R&D Policy > Advanced technology that clarifies complex systems

Advanced technology that clarifies complex systems

Example | System predicting abnormalities based on data from manufacturing operations

By means of artificial intelligence technology, machines can learn the normal behavior of devices operating at manufacturing facilities, so they can detect abnormal behavior at an early stage and notify supervisors.

Early-stage detection of abnormalities that are missed by existing alarm systems

The system learns from “big data” collected from operations and creates all kinds of monitoring models using a fuzzified neural network (FNN). Then, based on these models, it sends out estimated values. Based on the analysis of the relationships between the actual readings and the estimated values, it detects symptoms of trouble to prevent major problems from happening on the production line.

Fig. 1. Prediction and Detection of Abnormalities