How enterprise asset management is helping unravel mysteries of the universe at CERNSpeaking to Information Age at Infor's Inforum 2019 event in New Orleans, David Widegren, head of asset and maintenance management at CERN, discussed how it's using enterprise asset management (EAM) and engaging with AI/ML
The tools built and housed at CERN are used by the scientists to prove theories about the origins of the universe, and are some of the most complex machines ever built.
Its Large Hadron Collider, the largest and best-known collider in the accelerator complex, consists of millions of high tech components installed in a circular tunnel that is 16.7 miles (27km) long and situated 330 feet (100m) beneath the border between France and Switzerland.
However, according to David Widegren, head of asset and maintenance management at CERN, while CERN’s LHC, that runs underneath Geneva, qualifies as one of humankind’s most ambitious scientific contraptions, its operations involve challenges common to any other complex IT project.
With 22 member states, CERN is the world’s largest research centres for particle physics. On a given day, about 13,000 people could be working at the complex, including up to 12,000 visiting scientists from leading universities. The rest are looking after the nuts and bolts and day-to-day operations. With 700 buildings and a vast technical infrastructure, including tunnels, caverns, roads, parking lots, electricity, water, cooling and ventilation systems, access control, machine tools, lifting equipment, asset management is no mean feat.
EAM drives innovation
As such, EAM software is a big deal at CERN. For example, the organisation uses Infor EAM for the traceability of potentially radioactive equipment, ensuring that all equipment coming out of the accelerator complex is correctly measured for radiation and dealt with appropriately.
The centralisation of information through EAM, in this instance, empowers the radiation protection team to improve efficiency via making scheduled pickups, rather than sporadically visiting each exit point where equipment can leave the accelerator complex. Ultimately, this helps reduce the cost associated with sub-contractors in this process and gets the equipment back to end users faster, maximising the amount of science that gets done.
What does Infor’s process mining partnership with Signavio mean for automation?
According to Widegren, the rationale behind choosing Infor was their industry-specific offerings.
“We tried some of the package solutions but realised they were not doing the job for us,” he said. “We needed to have something that had certain functions and dedicated interfaces for very precise causes.”
Despite the operational benefits, according to Widegren, with the help of AI and ML, CERN is hoping EAM software can do more to improve the organisation in the future.
Indeed, significant challenges remain. For example:
- Unnecessary maintenance and downtime as a result of an inability to filter out the ‘noise’ of inconsequential device readings (airflow);
- Individual asset failures can severely impact accelerator complex uptime;
- Ageing installations but limited budget for equipment replacements and consolidation.
However, CERN has started utilising Infor’s Coleman AI Platform, alongside Infor OS – Data Lake, ION, Infor EAM, Infor LN. It looks like it’s paying dividends, according to Widegren, it’s “allowing us to quickly investigate real ML outcomes connected to EAM and operationalise results.
“What’s particularly interesting about the Coleman AI Platform is that it’s integrated with our EAM applications. This means we have an easier way of exploring the goldmine of information we’ve been storing.
Infor’s Inforum 2019: Embracing agile for ERP success
“With AI/ML, we are also able to explore the data in a better way. Instead of looking at what’s happening daily, we can go back through years of historical data and spot patterns, meaning we can move towards predictive maintenance.”
Looking ahead, CERN want to use their EAM systems to predict breakdowns in its accelerator complex by identifying patterns in alarm history and equipment dependencies and identify work order trends and patterns to optimise its maintenance.
According to Infor, with a more detailed analysis of individual devices, trends could also be correlated with additional aspects such as locations or age of devices.