A top 10 airport in the United States recently conducted a business case that combined artificial intelligence (AI) and Internet of things (IoT) devices to predict printer usage across the airport property. The business case project team for this airport conducted extensive research to make the case for the solution and identified information that can be leveraged by others in the airport industry who may be interested in addressing similar needs in their organization.

Note that this article will only reference information gained during the research conducted while defining the business case, as the airport organization had not yet approved the implementation of this business case. 

The solution outlined in this business case incorporated the following technologies:

• Internet of things: Connected devices that relay operational data or allow for remote management through the Internet. This business case used the data from connected boarding pass and baggage tag printers.
• Artificial intelligence: Computing functionality that can simulate human decisions based on design approaches to operational scenarios. The AI solution for this business case was used to support the machine learning functionality outlined in the business case.
• Machine learning: Functionality of an AI solution that analyzes data to identify and correlate relevant trends. The business case used machine learning solutions to identify printer usage trends and forecast future printer use numbers.

Case Study Explanation

In this case study, initiated from an effort to drive more data-integrated processes, the project team for this airport identified a solution that would leverage currently available data to predict the usage rates of common use baggage and boarding pass printer equipment within the airport. The business case outlined a plan to combine data from several sources related to passenger processing—including check-in time, flight information, and bag checking—to identify printer usage trends. A machine learning solution would be used to analyze these data sources, identify trends to predict consumption of bag tag and boarding pass paper stocks, and alert staff when printer resources were running low. The goal of the solution would be to reduce the need for staff to unnecessarily check various printer locations, freeing them up to do higher priority tasks and improve their cost-effectiveness. Machine learning and connected device data would enable this goal.

Upon initial investigation, it was noted that data from printers throughout the airport were being stored by the maintenance team as part of the vendor’s service contract; however, the data were never used. The project team then confirmed that printer usage data could be correlated with data from other airport processes or external sources to develop operational insights. Flight data and printer usage data could be analyzed through machine learning solutions developed by software engineering teams to identify what airport printer locations would face capacity strain and when their stocks were consumed.

Benefits Received

The project team determined that monitoring printer usage data had several tangible benefits to airport operations, including the following:

• Improved time and resource allocation
• New insight into business processes
• Cross-departmental use of data for process improvements
• Improved data visualization

Looking beyond revenue and cost savings can reveal several benefits to the entire airport operation. The data gathered from this process may be used for forecasting to drive efficiency in areas like the baggage handling system and staffing allocation. In the future, the project team hopes to integrate data across an increasing number of airport processes. As more data become available, more sophisticated solutions can be developed to drive further insights and improvements across the organization.

Areas Impacted

Following the research done for the business case, the project team identified several areas that may be impacted by deploying a solution like this. The following impacts identified by the project team have been categorized based on the impact areas detailed within this Publication.

Legal

Data ownership: An airport organization must thoroughly review its contracts to determine if there are any issues with data ownership. Depending on contract terms, an airport organization may need to coordinate further with a third party to access certain data streams.

Management contracts: Contracts may already be in place with established service-level agreements based on current processes. The parties managing the printer consumables may be from third-party organizations, adding an additional layer of coordination to project efforts. Project teams must take the time to define the responsible parties for system maintenance at the start of the project.

Technical Requirements

Necessary data integrations: To share data between systems, an interface needed to be developed to format the data in a way that was usable. Airport operators must identify and resolve any issues with accessing or connecting to the necessary data sources.

Process/Skill Set Changes

Training: Staff must be trained to effectively leverage the data in day-to-day processes. On-site staff must be trained to interpret the applicable consumable dashboard metrics and respond accordingly.

Data incorporation: Processes must be changed to only update printer stock when alerted by the system. The existing process called for routine checks of printer equipment, regardless of use. Operators must update their process to include data from connected printer sites that alert staff only when printer stock is low.

Budgetary

Budget for data access: In many cases, access to necessary data sources may only be available to a third party. An airport organization wanting to use this data may need to allocate funds for subscriptions to the equipment data feeds.

Budget for software development: Airport organizations must account for the budgeting of analytic software solutions to leverage the data gathered from connected devices.

Lessons Learned

The project team identified several new areas of information over the course of the project. While some areas were key to achieving the objective, others would prove to be useful lessons learned for future deployments. Following the project, several lessons learned were identified that should have been considered in more detail before the project began, including the following:

• Better collaboration with third-party vendors managing relevant systems
• Thorough understanding of airport data privacy and ownership
• Flexible processes to accommodate the added data
• Current airport devices may already be gathering process data

These factors impacted the efficiency of the project and became obstacles to address. While not applicable to all scenarios, airport operators should consider these factors to streamline their project efforts and increase the chances for a successful deployment.

Summary

This business case showcases the potential benefits that an airport organization can receive from currently available machine learning and IoT solutions. It also helps to illustrate how an organization can leverage currently available data to improve processes. However, when pursuing data analytics projects, it is important to understand the various stakeholders who may not be initially evident. As this kind of data becomes more valuable, parties may be more protective of the data they own, causing impacts to contracting and budgetary requirements.