Aviation Business News

PAM 2024: From data to action at AFI KLM E&M

Rob Stolk, project manager for predictive maintenance at Air France Industries KLM Engineering & Maintenance (AFI KLM E&M)
photo_camera Rob Stolk, project manager for predictive maintenance at Air France Industries KLM Engineering & Maintenance (AFI KLM E&M), on stage at Predictive Aircraft Maintenance (PAM) 2024 in Dublin

Rob Stolk, project manager for predictive maintenance at Air France Industries KLM Engineering & Maintenance (AFI KLM E&M) took to the stage at last week’s Predictive Aircraft Maintenance (PAM) Conference in Dublin to explain that by leveraging its end-to-end expertise as an airline and MRO organisation, many predictive maintenance models have been successfully adopted.

“Aircraft maintenance has evolved significantly with the integration of big data analytics,” said Stolk. “By examining billions of data points, engineers can uncover previously hidden trends and make informed decisions to enhance system reliability.”

Stolk’s presentation highlighted PROGNOS for Aircraft, the predictive maintenance suite developed in-house by AFI KLM E&M which is possible to predict a future breakdown or to schedule a maintenance operation beforehand. PROGNOS for Aircraft covers 35 aircraft systems across eight different types, including Airbus A320, A220, A330, A350 and Boeing 737, 747, 777 and 787.

Stolk’s presentation offered a compelling example of data-driven problem-solving through the case of the Air Cycle Machine (ACM). A hypothesis was proposed that ACM turbine 2’s underperformance stemmed from water ingestion.

“Detailed fleet analysis supported this theory, revealing recurring patterns of ACM speed drops upon the Environmental Control Valve’s activation,” explained Stolk. “By sharing these findings, collaborative investigations pinpointed the issue and proposed a solution – a drain hole modification that proved effective.”

Stolk explained that this collaborative approach yielded remarkable results, as evidenced by the improvement in ACM MTBUR metrics. “The fleet average MTBUR of 22,000 hours increased dramatically when environmental conditions were accounted for, with ‘dry’ MTBUR reaching 44,000 hours. These numbers underscore the transformative impact of targeted modifications informed by big data.”

The case study highlighted the importance of teamwork between data analysts, maintenance experts, and manufacturers. “The mantra ‘Data + People Together’ encapsulates the spirit of this initiative, emphasising that human collaboration is as critical as the technology itself,” concluded Stolk.

MORE FROM THE PREDICTIVE AIRCRAFT MAINTENANCE CONFERENCE

 

Sign In

Lost your password?