UC PhD student wins first prize at PHM 2014 Data Challenge

Seyed Mohammad Rezvanizaniani, a UC College of Engineering and Applied Science mechanical engineering PhD student, is the first place prizewinner of the data challenge that was featured at the 2014 Annual Conference of the Prognostics and Health Management Society. The conference was held September 29, 2014 to October 2, 2014 in Fort Worth, Texas.

Rezvanizaniani is projected to graduate in spring of 2015. He was born in Isfahan, Iran and received his bachelor's degree in Railway Engineering from Iran University of Science and Technology in 2002. Rezvanizaniani worked in Raja Rail Transportation Company as a maintenance engineer for more than six years in Tehran.

He received his master's degree in Maintenance and Management Engineering from Lulea University of Technology in Lulea, Sweden in 2008 and joined UC the center for Intelligent Maintenance Systems in 2009 to get his PhD in Mechanical engineering.

The Prognostics and Health Management Society (PHM) is a non-profit organization dedicated to the advancement of PHM as an engineering discipline. The PHM Society was incorporated in 2009. Each year PHM Society has an Annual Conference of the PHM Society, which is one of their biggest events of the year.

The Annual Conference of the PHM Society 2014 brought together the global community of PHM experts from industry, academia, and government in diverse application areas such as energy, aerospace, transportation, automotive, and industrial automation. The conference featured workshops, townhall meetings, hands-on demonstrations, a luminary session, a dedicated session on fielded systems, a doctoral symposium, and a full day of tutorials open to all registrants.

Rezvanizaniani summarizes the data challenge competition and states, “This year’s data challenge focused on asset health calculation, an industry problem found in remote monitoring and diagnostics for the unknown equipment at specific requested times. The dataset consisted of two years training dataset including maintenance records, usage and failures. In addition, one-year of testing data that included maintenance records and usage as well. The main objective was to determine if the equipment was going to fail within 3 days by using certain times within the testing data.”

Rezvanizaniani explained that GE provided the data for the competition. The contestants were unaware of the type of specific equipment it is. Rezvanizaniani created a data driven probabilistic model to define the high risk and low risk time intervals and consequently predict equipment failure. This model can improve the reliability and maintainability of GE equipment or any other industrial machine with same historical dataset. Predicting when a machine may fail enables production managers to schedule maintenance before that time to avoid shutdown, lost time, and machine replacement.

Rezvanizaniani has been an avid participant in PHM annual conferences since 2011. He states that his specific takeaways from this years competition was, “I developed a model based on training data to define the time intervals with high probability of failure for each asset using the reliability background and probabilistic risk assessment method.”

This is the fifth time a student from Intelligent Maintenance System center has won the PHM competition prize. They also won prizes in 2008, 2009 and 2011 as well.

Rezvanizaniani is currently at a local company doing an internship. He works at AMP Electric Vehicle. AMP designs and manufactures electric and hybrid trucks. He is also a member of the Society of Automotive Engineers international.

Select for more information about the 2014 Annual Conference of the PHM Society

Select to read about CEAS Professor Jay Lee, D.Sc., who was awarded the Jack Frarey Medal at the 68th Prevention Technology (MFPT) Society Annual