How do you use data to optimize airport decisions?

It's imperative for any data-driven organization to map out all possible data sources with periodical validation review cycles. Conventionally, passenger / baggage movements are captured through SITATEX messages, which are received by airports. Today, these messages are converted to meaningful data in systems, such as the airport management system (AMS). Once a system is in place to collect and store data, a reoccurring validation exercise to map out all data sources is extremely important, which is done by: 1. Mapping out all information that can be retrieved 2. Identifying all sources available to retrieve information 3. Setting the most credible source for each information 4. Validate its accuracy (data captured vs reality)

A key question about data is what data airports should collect. In my experience, “wait time”, the most expensive resource of all. Therefore, one relevant data sources are by “wait time” I mean the amount of time that passengers spend waiting in lines and queues at various touchpoints in the airport such as: Check-in lines - waiting to check bags and get boarding passes Security screening lines - waiting to pass through security checkpoints Customs & immigration - waiting in immigration/customs queues Boarding lines - waiting to board the aircraft at the gate Baggage claim - waiting for checked bags to arrive at carousels, etc.

Yes definitely. Data Analytics can not only help the airport authorities to understand the current trends, it can also help them to forecast load during off peak or on peak seasons, holidays versus business travel, maximum peak hours etc. With the help of such forecasts, they can perform efficient resource management as well as streamline operations.

What is data analysis in simple terms? Translating data into meaningful information. By analyzing data, we can measure the airport's performance, current capacity utilization, areas that need improvement and even celebrate success for a job well done (something that might even be missed). Not only does it benefit us in covering historical data, it also allows us to identify trends in order to establish data-oriented projections. These projections could forecast the number of passengers to prepare an airport's capacity beforehand, plan resources (manpower) based on demand for improved optimization, and anticipate irregularities in order to enhance the airport's operational readiness / resiliency capabilities.

TO help airport management to undrstand patterns, trends, correlations, etc. I propose using machine learning algorithms leveraging artificial neural networks to model wait time patterns across different times of day, days of week, and peak travel seasons and make accurate predictions about bottlenecks, as well as creating highly user-friendly interactive data simulations incorporating these predictive models to allow managers to test a wide range of operational changes in a risk-free environment and select optimized solutions, in order to continually maximize operational efficiency and positively enhance the customer experience in impactful lasting ways.

In my experience this is the part where things get to get really exciting. All those hours of work you have spent in sifting, mining and cleaning the data finally starts to take shape in front of your eyes into a story that can speak for itself. This is where everyone from airport staff to customers as well as management starts to “see” what’s happening where and the core areas of improvement.

To create a data visualization project for an airport, you may need to consider the following factors: The type and purpose of the data visualization: What kind of data do you want to visualize? What questions do you want to answer? What insights do you want to convey? The data sources and quality: Where will you get the data from? How reliable and accurate is the data? How often is the data updated? How will you clean, process, and transform the data? How will you choose the appropriate visual elements, such as colors, shapes, sizes, labels, legends, etc.? How will you ensure the data visualization is clear, effective, and user-friendly?

Data governance is a continuous effort to validate the integrity, credibility and accuracy of vital data. In airports, this is normally done through repeated data validation exercises, where all data sources are mapped out. It's important to compare data with real figures, these can be obtained through targeted spot checks in order to not completely rely on a system when it comes to checking its own accuracy. One of the most important factors here is transparency. Accuracy checks should be done periodically and being transparent with the results will help improve performance while building trust / credibility to the data.

Privacy and security of passenger data are very important. There exist various methods and technologies to ensure that the data collected, process, and visualization are protected and compliant with the relevant laws and regulations. Some of these methods and technologies are: Data anonymization and encryption. Data access and control: Secure authentication and authorization mechanisms, such as passwords, tokens, biometrics, etc. Data storage and transmission.

Changing one's mindset is never easy, cultural change must come naturally. The first and most important step to creating a data-oriented culture, is to help people understand "why" it's needed. How can data help them? It could justify their business cases, measure their outstanding performance, forecast the need for additional resources, etc. Cultural change will not happen by presenting a slide to an audience about the desired change. True change is when people buy-in to an idea. Once they believe in the idea and how it will help them, change is inevitable and natural rather than being forced.

There are many common challenges with data at airports. The first is archiving data, given that there will be a large amount of data continuously stored and sustaining high storages may require additional investments. In order to reduce the load on systems, users may have limited access to historical data (e.g. last 3 months) and prolonged periods are made available upon request. Some airports have over 20 years worth of data, which is reducing the system's performance to the point that it's unusable. It's important to create a separate archive (data depository) and not link it with users that require it for operational use. Such archives can be dedicated to analysts. With time, meaningful analysis can be shared and easily accessible.