TRANSFORMING TRANSPORT AIRPORTS ATHENS
PILOT : SMART PASSENGER FLOW AT ATHENS AIRPORT
BIG DATA VALUE IN MOBILITY AND LOGISTICS
LOCATION ATHENS INTERNATIONAL AIRPORT
CONTACT NAME JUAN FRANCISCO GARCÍA LÓPEZ
PILOT PRODUCT DESCRIPTION
Year 2017 was one of strong momentum for the global aviation industry, characterized by low oil prices, efficiency of aircraft capacity utilization and fleet growth. The combination of the above resulted in a very prosperous year for airlines’ profitability, with attractive fares, which in turn supported a strong growth of passenger traffic. According to the Athens International Airport (AIA) annual Airport Statistics Handbook1 published recently, 2017 has clearly proven to be a record-breaking year for AIA in terms of annual passenger growth.
Increasing by 1.7 million compared to the previous year, passengers grew by a significant 8.6 per cent and amounted to 21.74 million approximately. Cumulatively since 2013, AIA has experienced a growth of approximately 73 per cent, reaching an aggregate of 21.7 million passengers from 13.7 million. This is a result of the ‘spring effect’ following the recovery from the financial crisis together with the global growth of the aviation industry. Such a phenomenal growth allows little time for airport planners to compile, submit and approve passenger capacity–related investments, such as civil engineering–related projects, ie terminal building expansions or the building of entire new ones. Such projects, in general take a period of three to five years to be implemented — and in the case of AIA, where passenger numbers rapidly increased, even when those investments are realized, the expected result would not solve the airport’s capacity constraints. Other more cost-effective and efficient (intelligent) solutions that would allow to ‘sweat’ the current assets or even revolutionize their use and allow for process optimizations, are investigated in an effort to achieve passenger throughput maximization for the existing passenger terminal facilities. Aircraft turnaround optimization combined with the minimization of passenger process–related delays would postpone relevant investments and warrant the required preparation period for the design and realization of such investments.
In this respect, the analysis of passenger behaviors, the segmentation of passenger traffic to the constituting categories, the demand for airport facilities and the forecasting of the passenger numbers not only helps to achieve understanding of the passenger process but also helps to propose actionable strategies that will significantly increase the passenger throughput as well as the passenger experience.
The way to proceed to this analysis is to employ data analytics regarding the passenger process and create synergies with other airport stakeholders, such as airlines, ground handling agents and concessionaires, so as to build the relevant profiles that will allow for the creation of the descriptive models to help understand and — at a subsequent step — predict passenger behaviors.
OBJETIVE OF THE PILOT
The two main objectives and sub-objectives of the Smart Passenger Flow pilot are the following:
Obj.1: Operation Management Predictive Optimization Module
By means of a real time operational module, the pilot will exploit predictive analytics with Passenger Flow data obtained in real time from airport and airline systems. This module will facilitate proactive decision making in real time whenever there is any disruption over the initial plan. The initial targets to be covered with this module are explained in the following sub-objectives:
- Sub Obj.1.1: To provide a more refined prediction of delays in departure flights according to late passengers.
- Sub Obj. 1.2: Reduce the number of passenger missing connections.
- Sub Obj. 1.3: To improve the efficiency of passenger processing stations.
- Sub Obj. 1.4: To predict possible time savings in the turnaround process.
Obj. 2: Descriptive passenger behaviour system
- Sub Obj.2.1: Obtain insight on how passenger behave along their journey, especially within the airport terminal to enable customized services/offers, increasing passenger satisfaction and non-aeronautical revenue.
The main expected results of this pilot are:
- Identification of passenger behavioural patterns so as to make informed decision on how to optimise operations.
- Discovery of passenger’s segments and anonymous demographics so as to better address passenger needs.
- Early identification of operational anomalies (eg delays) through the analysis of historical data, machine learning and patterns and trends discovery.
- Prediction of passenger traffic and demand of airport services (eg security) so as to meet service levels.
Allow the creation of new business models based on data- driven decision making in retailing.
KPIS AND METRICS
The main KPIs of the pilot are:
- Optimization of the number of security lanes required to manage the passenger demand, avoiding underused infrastructure and bottlenecks.
- Reduction of staff cost due to the optimization of security lanes
- Reduction of airport energy consumption, due to the optimization of airport resources, and therefore, reduction of CO2 emissions related with the generation of energy