The temperature was rising in Geneva, Switzerland, during June, as maintenance personnel and leaders from across the globe touched down to attend a one-day Infor Enterprise Asset Management (EAM) European Summit at The Globe of Science and Innovation, an emblem for CERN.
Billed as an exclusive experience to connect with other companies, share challenges, and gain new insights into the digital transformation of, and best practices for, asset management, cloud, mobility, and predictive maintenance, around 180 delegates filled the venue.
Among the speakers was David Widegren, head of asset and maintenance management at CERN, who is in charge of managing the lifecycle of the organisation’s physical components and assets. He explained more about CERN and its asset management platform, as well as future strategies surrounding artificial intelligence (AI) and machine learning.
CERN is the world’s largest research centre for particle physics, with 12,000 visiting scientists, 2,500 students and 2,600 staff at the campus on any given day. Made up of 23 European states and over 70 additional collaborating companies, its mission is to provide scientists with the tools to study the building blocks of matter and the origins of the universe.
How does it do this? By building and operating some of the most complex machines ever built; huge particle accelerators. The most notable accelerator is the Large Hadron Collider (LHC) – the largest collider in the accelerator complex. It consists of millions of high-tech components installed in a 16.7-mile circular tunnel, situated 100m beneath the border between France and Switzerland.
CERN has been an Infor customer for the last 30 years, with Infor EAM, an asset management software that can help digitise and optimise maintenance operations, being used to track the design, operation and maintenance of key pieces of equipment in the accelerator complex. This ranges from everyday infrastructure, such as elevators, to the magnets of the LHC.
The numbers are impressive. CERN currently manages some 2.4 million pieces of equipment on the system, and has more than 200,000 work orders a year. On top of this, it averages 5,000 checklist items per day and has 92,000 part references to hand.
Integrating Infor EAM has given control centre staff and maintenance personnel a clearer view of maintenance history and previous problems. Operators can use information to diagnose a malfunction and ensure that corrective maintenance occurs quickly. Visibility into maintenance history also aids in optimising when and how much preventative maintenance should be performed.
“We have two keys to success,” Widegren said to delegates. “Firstly, we have integrated Infor EAM with our other tools, and secondly, we have simplified our user interfaces for specific tasks.” Infor EAM is the central repository for physical assets and is used via different user interfaces. Whereas the majority of people involved in maintenance activities at CERN are using Infor EAM or EAM Light user interfaces, others might be using it via TREC (radioactive traceability) or MTF (manufacturing test folder), which are also simplified web interfaces created for specific tasks on top of Infor EAM. There is, in other words, one single database with multiple user interfaces (www.is.gd/kokelo).
CERN’s assets often have a lifecycle of more than 50 years. This lifecycle, combined with personnel turnover, makes asset and intervention documentation a must.
AI & FUTURE STRATEGIES
AI and machine learning efforts at CERN stretch back to the 1990s, being used within physics research to analyse data and make predictions. However, AI and machine learning “can also be used in other areas”, Widegren explained.
“For some time now, we have been using machine learning on control systems and operations,” he said. “This is assisting our controlroom operators in taking more informative decisions, while proposing options and likely faults diagnosis.”
CERN is also looking to do “a similar thing” around asset and maintenance management, allowing it to move towards predictive maintenance. Widegren explained that, for a long time now, CERN has collected a lot of physical asset data without exploring its full potential. “By combining and correlating data from multiple sources, we might discover hidden patterns and dependencies, as well as identify root causes in complex systems,” he added.
Another exciting development surrounds an AI pilot project between CERN and Infor Coleman, Infor’s complete artificial intelligence solution that can be used to predict events and alert users to activities that require attention. They are currently starting up a pilot project where several ‘use cases’ will be looked at, with the first surrounding predictive maintenance for smoke detectors.
“We have some 500 air sampling detectors [on the accelerator complex],” explained Widegren. “They don’t just detect smoke, but also air flow and particles in the air. Preventative maintenance is currently carried out every three months, so the goal is to optimise maintenance intervals and avoid false alarms.
“The two teams are hoping to achieve this by correlating air flow and smoke detection measurements within asset data, as well as the alarms history.”
Widegren added that with more tightly integrated engineering tools and improved linking of data, CERN is also aiming to manage a digital twin of its installations.
He concluded the presentation by reiterating three points: CERN is accelerating its efforts in extending the use of Infor EAM; AI and machine learning capabilities will enable CERN to explore asset data better and ease towards predictive maintenance; and by managing a digital twin of its installations in an integrated engineering platform, data will become more accessible, operations and maintenance more efficient, and CERN will be better prepared for the digital era.
BOX OUT: Up close with the Synchrocyclotron
Following the EAM Summit, delegates were given a guided tour of CERN’s first accelerator, the Synchrocyclotron (SC), which was powered down just as Infor started work at CERN (main picture).
It all began in 1949, when Europe was recovering from World War Two. Many notable and eminent scientists had left for the United States and it was said that European research was no longer world-class. To reconstruct European science, French physicist Louis de Broglie proposed the creation of a European laboratory that would act as a centre of excellence in physics and a motor for peace.
His idea was taken up by American Nobel prize winner Isidor Isaac Rabi, who together with pioneer Pierre Auger and Italian physicist Edoardo Amaldi, convinced UNESCO to adopt such a process. Eighteen months later, 12 European nations (see below) formally agreed to create a ‘Conseil Européen pour la Recherche Nucléaire’ (European Council for Nuclear Research) and the acronym CERN was born.
In the first sessions of the new council, important decisions were taken. Geneva in Switzerland was chosen as the seat because of its central location in Europe. Two accelerators were also put forward – the SC and a much larger machine, the Proton Synchrotron. A new convention was also laid out to define CERN’s goals: its research would have no concern with military requirements and all research would be made public. The CERN convention was signed by 12 countries in 1953 – Belgium, Denmark, France, the Federal Republic of Germany, Greece, Italy, the Netherlands, Norway, Sweden, Switzerland, the United Kingdom and Yugoslavia – and CERN finally came into existence.
THE SC BUILD
In 1954, on a site in Meyrin, a village near Geneva, work on the new laboratory began. Within a year, the farmland was transformed into a large complex of workshops, offices and buildings, to house the new accelerators. The five-metre-thick walls of the SC building emerged in 1955, while the parts of the accelerators were manufactured across Europe. The SC magnet coils required particular transport care.
Once all components arrived, they were assembled with great precision. The SC began to emerge piece by piece. Some 54 elements were put together for the magnet frame – a total of 2,500 tonnes of steel. The two 7.2m diameter magnet coils were installed next, which had a current of 1,800 amps that generated a magnetic field of two Tesla.
A rectangular vacuum tube was then fixed inside the magnet, while a system of large vacuum pumps were installed to evacuate the 25m3 chamber, allowing protons to circulate unimpeded. This was followed by the installation of two D-shaped electrodes inside the vacuum chamber. Together with these electrodes, the radio frequency generator could create an electric field and accelerate protons from a source inserted in the centre of the vacuum chamber.
COMING TO LIFE
In summer 1957, the SC was finally ready, and CERN’s first accelerator came to life with the purpose of producing and studying new particles. The new machine could accelerate protons to 80% the speed of light, producing millions of new particles when the protons collided, giving scientists the opportunity to make systematic measures. Operation of the SC required a sequence of actions:
● Pumps would extract the air from the vacuum chamber so that protons could not collide with gas molecules during acceleration
● In the proton source, hydrogen gas was ionised, and a cloud of protons injected into the middle of the SC
● The accelerator would then make use of electric and magnetic fields; protons have a positive charge and would be drawn towards the negative electrode as they traversed the gap between the electrodes
● The magnetic field would force them to follow a circular trajectory and return to the gap after one half turn
● The radio frequency generator would then reverse the polarity between the two electrodes. The protons, now attracted to the opposite electrode, would begin to gain more energy. This process would be repeated over and over, whipping around faster with every half turn. The radius of their path would increase
● After more than 100,000 turns, they would reach an energy of 600 million electron volts (MeV) and move at 80% the speed of light.
Young scientists from all over Europe began to flock to Geneva, among them Maria and Giuseppe Fidecaro. Giuseppe wanted to study a short-lived particle called the pion and, in 1958, he and his team set up an experiment. The pions were stocked in an apparatus designed to study their subsequent decay. Just a few hours into the experiment, the first pictures showed evidence of this rare decay.
“It was something exceptional,” Giuseppe said. “Because something was seen for the first time that nobody had seen before.” This first important discovery spread CERN’s name around the world and, over the following years, scientists at the SC continued to make many more important measurements on particles, atoms and nuclei.
BIGGER AND BIGGER
In 1967, a new idea took shape called Isolde (the Isotope Separator On-Line DEvice). Protons from the SC would collide with target nuclei, which would split into short lived fragments and be scrutinised in experiments. The study of nuclei helped to understand how heavy elements are produced.
The SC supplied the Isolde experiments with a large variety of particles over almost 25 years. Then, in December 1990, the SC finally accelerated its last beams. Over the years, bigger and bigger accelerators were built to progress research, including the Proton Synchrotron (1959), the Intersecting Storage Rings (1971), the Super Proton Synchrotron (1976), the Large Electron Positron Collider (1989), and the LHC (2008).
However, the SC, with 33 years of service, will always sit in the history books as CERN’s first accelerator.