PARC Launches Condition-Based Maintenance Platform Offering Insights into System Health, Safety, and Performance

Clients and Partners Include BAE Systems, General Motors, Hitachi, LG Chem Power, IHI, Xerox


Palo Alto, CA, June 27, 2016 (GLOBE NEWSWIRE) -- PARC, a Xerox company, announced its Condition-Based Maintenance (CBM) platform, a suite of innovative software and hardware technologies that work together to offer insights into system health, safety, and performance. PARC’s sensing, modeling, diagnostics, machine learning, predictive analytics, rapid prototyping, and artificial intelligence, have been being developed and perfected at the renowned research facility for more than a decade.

PARC worked with IHI Corporation, a diversified manufacturer producing a wide range of high-end products such as jet engines, rockets, ships, storage, processing plants, and industrial machinery, to demonstrate and deliver its “Plantrol" planning and recommendation software for high-value manufacturing plants for rapid re-configuration across a diverse set of scenarios. “The key to our success was PARC’s model-based approach, which allowed us to quickly repurpose a solution initially developed for one domain to other complex ones,” said Koji Tanaka, Director, Products Development Center, IHI Corporation. “Plantrol was successfully adapted to different types of IHI’s manufacturing plants. PARC’s system would help us quickly understand robust reconfiguration options using failures in real-world scenarios and, importantly, was scalable across our various plants.”

The reliability and uptime of a myriad of infrastructure systems is becoming increasingly critical. As services budgets continue to be stretched, maintenance, operations, manufacturing, and design teams are under tremendous pressure to maximize system life and utilization without compromising safety and operational uptime. Major organizations are already taking advantage of PARC’s innovations, including BAE Systems, General Motors, Hitatchi, LG Chem Power, IHI, and Xerox, among others. Refrigeration systems, elevators, railways, transportation infrastructure, energy storage, and smart manufacturing are just a few industries and systems being disrupted as a result of PARC’s CBM technology.

“With today’s practice of reactive fail-and-fix or schedule-driven preventative maintenance, it’s inevitable that some systems will fail, often resulting in serious accidents,” said Ajay Raghavan, PARC Research Area Manager. “PARC’s CBM is a model-based approach, enabling higher than 90% accuracy and negligible false alarm rates, arming our customers with actionable data for informed deployment. It gives them the ability to deeply understand their systems to smartly manage and keep them in top shape, or be able to know when to take them offline to fix them before any unfortunate events.”

Our CBM platform components are driven by PARC’s deep, interdisciplinary expertise in system sciences and engineering. We start with low-cost embeddable sensors that can be placed into the heart of systems in a non-invasive manner to directly monitor system state. We augment the sensor data with our system models that we can automatically augment with a range of faults that can be anticipated for the model. This allows us to develop a deep understanding of how the system works and fails, enabling us to quickly develop informed failure-sensitive algorithms to detect abnormal system operation.

This understanding feeds into model-based diagnostics that PARC is renowned for pioneering, allowing for highly accurate classification of system faults and isolating vulnerable subsystems that lead to system failures. Additionally, PARC CBM components feed into prognostics algorithms that can predict failure points under different expected load scenarios. Finally, PARC combines this information along with specific response suggestions into condensed recommendation algorithms.

CBM is a step towards PARC’s broader mission to enable self-aware, self-adaptive systems. This is driven by PARC’s strong foundation of model-aware methods to facilitate a potent first-principle representation of systems. This gives the system a deeper “self-awareness” and makes it easier to scale up to other conditions and climates and enables other system management functions. In contrast, conventional machine learning approaches that are only data-driven are much harder to scale and will not go beyond what you specifically train the machine to do. We are working toward enabling a paradigm where systems manage themselves autonomously with minimal to no human intervention, taking system uptime, performance, maintenance, and reliability to the next level.

About PARC 
PARC, a Xerox company, is in the Business of Breakthroughs®. Practicing open innovation, we provide custom R&D services, technology, expertise, best practices, and intellectual property to Fortune 500 and Global 1000 companies, startups, and government agencies and partners. We create new business options, accelerate time to market, augment internal capabilities, and reduce risk for our clients. Since its inception, PARC has pioneered many technology platforms – from the Ethernet and laser printing to the GUI and ubiquitous computing – and has enabled the creation of many industries. Incorporated as an independent, wholly owned subsidiary of Xerox in 2002, PARC today continues the research that enables breakthroughs for our clients' businesses.


            

Contact Data