Problem predictor

US researchers are developing intelligent computer software capable of predicting when systems are about to break down or need special attention.


Researchers at the Air Force Research Laboratory are developing intelligent computer software capable of predicting when systems are about to break down or need special attention.


The software, called Prognosis, has already been used to improve the reliability of high-power advanced chemical lasers.


Prognosis can predict conditions, circumstances and faults. It is being developed by Victor Stone, a computer engineer at the laboratory’s Directed Energy Directorate, and Dr. Mo Jamshidi, a professor at the University of New Mexico.


Prognosis comprises a variety of neural network systems, fuzzy logic, pattern recognition, data mining, classical expert systems, and hierarchical and hybrid software tools. No one tool works in every situation but the most appropriate tool can be selected for a particular application or many of them can be used.


Stone and Jamshidi are working with technicians at Intel Corp in Rio Rancho, NM to tailor the Prognosis system to Intel’s needs. They believe that if the software was installed on just one portion of Intel’s production line, the savings could be as much as $7 million a year.