In August 2017, a glitch in the baggage handling system at Toronto Pearson International Airport’s Terminal 3 caused flight delays of several hours, resulting in severe reputational damage for the airport. Many passengers took to social media to vent their anger and frustration with the airport management authority.
A sound maintenance program is critical for extending the life of airport facilities and keeping the airport as safe and efficient as possible. The lack of predictive maintenance efforts at airports can not only result in premature failure of infrastructure and unplanned downtime, but also lead to an emergency situation with major cost implications.
What is Predictive Maintenance (PM)?
The tremendous technological advancements in sensor and communication technologies over the last decade have made it possible to monitor assets in near real time. Asset performance data analysis can help predict potential failures so that timely actions can be initiated to avoid unplanned downtime. This approach of Predictive Maintenance leveraging IoT and data analytics ensures a minimal element of uncertainty in airport operations.
Predictive maintenance solution components
Assets: Airport assets can be broadly categorized into airside assets and landside assets. Based on the role these assets play in airport operations, they can be classified as critical or non-critical assets.
- Airside Assets (Aerobridges, passenger coaches, cargo vehicles, refueling trucks, etc.)
- Landside Assets (Baggage handling systems, elevators and escalators, building facilities like lighting, HVAC, security and surveillance, power back-up systems, etc.)
Sensors and gateways: Assets are evaluated for their sensing and connectivity capabilities, i.e., their ability to measure and publish data on their health conditions. The asset data is streamed through an edge gateway periodically via IT infrastructure to an on-premise or cloud hosted remote server. Assets which don’t have native support for such capabilities are enhanced either through re-engineering or through add-on external sensors and networking modules.
Enterprise Asset Management platform:An Enterprise Asset Management (EAM) platform plays the pivotal role of acting as a central data repository by ingesting asset performance data and aggregating data from other airport systems. When advanced data analytics engines with machine learning models are executed over this reference asset data set, they offer critical insight to the airport managers on real time asset health, an asset’s propensity to fail, and related maintenance requirements to avoid asset failures and outages. These insights can further trigger alerts and notifications to concerned technicians, and lead to automated work orders for proactive maintenance jobs and to manage the complete operational workflow.
Traditionally, Enterprise Asset Management platforms have been offered as an on-premise solution only, leveraging historical data to offer limited descriptive analytics. However, going forward, cloud-based options and hybrid (on-premise and cloud-based) options with seamless support for advanced data analytics and machine learning capabilities are expected to be the norm.
Typical solution architecture