Translated by
2019/04/23 10:40:02

"Air mathematics". Big Data in the world of civil aviation

Partner of the overview
Partner of the overview
In 2017 ― 2018 digital transformation, i.e. gradation of all industries on modern technologies becomes the main trend of economy. Also the aircraft does not stand aside. Big Data technologies which are capable to increase their operational reliability and efficiency become an important basis of development of airlines.


Technologies <!--LINK 0:1--> become an important basis of development of airlines
Technologies Big Data become an important basis of development of airlines


According to the forecast of analysts, 67% of the companies from the space industry implement projects on the basis of Big Data, another 10% plan such projects. As for airlines, here project implementation for February, 2019 is declared at 44% of the companies, and plans for such projects announced 25%.

These are results of a research which was conducted in December, 2017 by FlightGlobal company concerning Big Data role for the space enterprises and airlines. Analysts also found out opinion on sharing of data on a status of airplanes with the producers and the companies performing repair and the maintenance (M)[1]. 300 professionals from the space and aviation industry participated in a research. Most of them are sure that Big Data technologies are capable to increase operational reliability and efficiency of airlines.

Infographics with results of a research
Infographics with results of a research

About a half of respondents answered that their companies use data arrays about a status of airplanes that helps them to make more verified decisions. In the short term the share of such companies will grow to 75%.

Sharing of data with OEM/MRO still remains problematic. However 38% of airlines believe that such model can provide them considerable business benefits.

According to data from the overview of Connected Aircraft Honeywell company[2] published in May, 2018], 47% of the polled airlines are going to spend for the purpose of connection of aircrafts to Network up to 1 million dollars within the next year for each airplane operated by them. Most these companies is going to keep within the amounts from 0.1 to 0.5 million dollars. However, in five-year perspective of 38% of airlines announced investments already in the amount of 1-10 million dollars on each airplane.

Till February, 2019 at investment by airlines in adjacent to aircraft of technology (connected technologies) it was talked, first of all, of providing satellite communication and Wi-Fi. Now the companies are ready to benefit by those data which they can receive by use of the equipment directly onboard airplanes. For example, such data can provide them economy in the amount of 1% of the consumed fuel that 50,000 dollars aboard the plane in a year are equivalent, analysts of Honeywell counted.

Big Data use by foreign airlines

Big Data technologies are applied to accomplishment of a number of tasks in the field of civil aviation. In this chapter we will dwell upon the main directions of use of Big Data in aircraft in a number of foreign countries. First of all it is repair and maintenance, ensuring economy of fuel, creation of digital doubles, optimization of operating activities (including forecasting of delays of runs), formation of personal offers for passengers, etc.

Big Data technologies apply airlines worldwide
Big Data technologies apply airlines worldwide

Big Data and maintenance of the flight validity of aircrafts

The maintenance (M) and ship repair will be one of such priority areas in the short term. So, 88% of respondents of analytical researches expect that they will be able to get the maximum advantages of use of technologies in this sphere. Maintenance and repair considerably advance all other spheres on importance. The analysis of Big Data and predictive repairs in aircraft show the efficiency and prove in practice that connected technologies work.

After TO and repair respondents expect the advantages of implementation connected in the field of technologies of piloting including optimization of consumption of fuel and turn-around time for airplanes and also service of passengers.

So, in a research "Sky High Economics: Evaluating the Economic Benefits of Connected Airline Operations"[3] analysts note that the connected airplanes can transfer data to a cloud or on land servers where these data can be analyzed using the Big Data Analytics tools. Thanks to it airlines can reveal, for example, faults before they become large problems. The acquired information can be used for adoption of more verified solutions and reduction of expensive idle time of airplanes (Aircraft on Ground).

Besides, with the advent of forecast modeling (predictive modeling) replacement of parts which based on the analysis are defined as the requiring replacements became possible before they failed, namely during planned works on repair and TO. All this promotes cost reduction, increases flight safety.

Digit Twins. What is it?

Use of so-called "digital doubles" ("digital twins") is also closely connected with a subject of predictive (pro-active) repairs. However in difference, for example, from the oil and gas industry where TsD are already used by a number of the large companies in practice, in the aviation industry this subject is still more discussed at the level of experts and analysts.

Experts of the aviation industry in 2019 began to advance a subject of use "digital twins" actively: the management of the Swedish company IFS, software developer for corporate clients, including from the aviation industry, said in April, 2018 that one of technology innovations capable to help airlines to provide effective operation of courts at simultaneous expense reduction on TO and repairs, "digital doubles"[4] are]. Digital doubles ― it is the virtual remarks of physical assets capable to show to engineers on the earth operation of the engine while the airplane is airborne. To make it possible, engineers set at a design stage and productions of the engine thousands of points of data collection. Then they are used for creation of digital model which traces and controls the engine in real time, providing necessary information for all its lifecycle, for example, temperature, pressure and a consumption of air.

GE helped to develop the digital double for the landing device of the airplane. Sensors were placed on parts of the chassis, to the most subject breakdowns. In real time such data as pressure and temperature, were transferred to specialists, helping to predict failures in work or the remaining service life. These data were compared to data of the digital double who was exposed to similar loadings.

According to IDC, the companies investing in TsD are capable of 30% to reduce time required for implementation of critical activities including TO. Analysts expect that in 2019 the technology will become more mature and will provide additional benefits for users.


The MarketsandMarkets analytical company published in May, 2017 the forecast of the market of the chat-bots Chatbots Market Growing at a CAGR of[5]. According to data of analysts, annual average growth rates of the market between 2016 and 2021 will be 35.2%. In terms of money the market will grow from 703.3 million dollars in 2016 to 3.172 billion dollars in 2021. A main objective of implementation of chat-bots in the aviation industry is desire of the companies better to understand a consumer behavior of their services and goods.

Besides, Big Data-solutions begin to use in the aviation industry and for the solution of other tasks, for example, for ensuring aviation security based on the analysis of arrays of historical data or, for example, that has something in common with increase in level of client service, for process optimization, connected with catering services onboard, and trackings of baggage.

Examples of implementation

Let's review examples of implementation of the solutions Big Data by foreign airlines. The main focus is placed on large-scale projects in recent years and also on those business cases in which data on quantitative or qualitative effects of implementation are provided. These business cases belong to perspective spheres which were described above.

Repair and maintenance using Big Data

The British low-cost airline EasyJet, the operator of the Airbus A319 airplanes, is going to implement about 50 different algorithms for predictive TO and repair onboard the courts. The technical guidance of the company said in the middle of June, 2018 that the works on all fleet should be completed by the end of 2019[6].

EasyJet is going to implement about 50 different algorithms for predictive TO and repair onboard the courts
EasyJet is going to implement about 50 different algorithms for predictive TO and repair onboard the courts

For implementation of model of forecast repair and TO two solutions ― an onboard data exchange system on flight transactions and TO (FOMAX) and off-board tools will be used for the analysis of aviation data of Skywise. FOMAX, the server from Rockwell Collins, collects data on TO and performance of the airplane, in the automatic mode sends them to engineers and technicians. SkyWise working at a cloud platform provides the analysis of different data, was developed jointly by Airbus and Palantir Technologies companies.

The FOMAX system obtains all data from located onboard the FDIMU system (flight data interface management unit). FOMAX has functionality of the 4G-router: after landing of the vessel all data with the help 4G of Gatelink-antennas are transferred to the analytical Skywise platform and analyzed by specialists of Airbus. Specialists of Airbus independently developed the special models capable to predict emergence of system malfunctions for the analysis. After the carried-out analysis its results are transported to specialists of EasyJet who already independently make decisions on need of carrying out predictive TO or repair. Using the acquired information, specialists of airline can create algorithms which will allow to predict in the future emergence of this or that malfunction on any of airplanes.

A320 airplanes with FOMAX are capable to collect more than 24.000 parameters, i.e. to provide 100% collection of information from systems and components of airplanes. Airplanes without FOMAX collect 400 parameters, i.e. 2% of available information.

Meanwhile, the Delta Air Lines airline reported in April, 2018 that it is going to adapt the system of maintenance with predictive functionality for airplanes of the last generation which generate considerably large volumes of data, than the previous vessels (the airline prepares for acceptance of the next models of airplanes, such as Airbus A350 and Bombardier CS100)[7].

According to statements of the management of Delta, the used program of forecast TO helps airline to reduce number of failures in work considerably: For the last 12 months use of pro-active TO helped to avoid 1.200 delays with a departure or cancellings of runs.

The program uses the data arriving from the different systems, such as, for example, Aircraft Health Management from Boeing, from the Airbus and GE systems. At the same time, the basis of plane fleet of airline is formed by airplanes which were developed still until as extraction and data analysis of must-have steel - function. After the analysis of data retrieveds the program develops recommendations about replacement of parts and mechanisms. According to specialists of Delta Air Lines, the used program shows the 95% level of accuracy regarding the recommendations as necessary of replacement of parts.

The Honeywell company in May, 2017 together with Cathay Pacific tested use of the solutions Big Data Analytics by airplanes of the Hong Kong[8]. The tested solution ― ― is focused by GoDirect Maintenance Service on use of methods of the forecast analysis regarding maintenance of aircrafts.

Provided to Cathay Pacific for testing of the solution the A330 airplanes. Data from the equipment of airplanes were transferred on the existing channels to divisions of airline and also to the providers rendering services. Thanks to it, service staff on the earth could be prepared for potential works on TO in advance and also in advance order spare parts, without waiting for approach of critical situation.

During testing it was succeeded to reduce the number of breakdowns of the equipment of the airplane by 35% that allowed to reduce costs for TO, to reduce quantity of delays of a departure and a downtime of airplanes. Accuracy of prediction of breakdowns was 99%, declared the management of Honeywell.

In June, 2017 the companies signed the agreement on the implementing solution GoDirect Connected Maintenance, the using Big Data and sensors for forecasting of faults, by all Airbus A330 airplanes of airline. According to the contract, the GoDirect Connected Maintenance systems will be installed by more than 60 Cathay Pacific airplanes. According to Honeywell, the Cathay Pacific company considers the possibility of use of a system also by the Boeing B777 airplanes.

East neighbors also do not stand aside. In the fall of 2017 the Korean Air company began to use technologies based on artificial intelligence for support of the specialists who are carrying out by TO[9]: Korean Air uses machine learning technology (Machine Learning) for processing of data bulks which are generated by courts of the company. The solution relies on the Watson platform from IBM on the basis of artificial intelligence and is capable to read the last records concerning TO, made in a natural language (natural language) and also to process the data arriving from the airplane and technical specialists.

For the solution of similar tasks Korean Air uses technology of the content analysis based on WatsonExplorer and Natural Language Understanding (NLU) that helps to process the structured and unstructured data. It allows to reduce time required for determination of the reasons of possible fault in the future or the existing problem for 90% in comparison with traditionally used methods.

Examples of cost reduction of fuel and emissions upon transition to Big Data technologies

In 2016 the American airline Southwest Airlines decided to test the analytical platform which allowed to reduce the second in value cost item ― the expenses on fuel making from 4 to 6 billion dollars annually[10]. Initially cost forecasting on fuel in Southwest Airlines was performed based on information from several systems, including Ariba, the environment of management of fuel consumption of Allegro and also from own corporate storage of historical data of the company. All these data came down then in one bulky table. Specialists monthly generated 1.200 forecasts of need for fuel. The financial analyst spent a month three days of work for drawing up these forecasts which, besides, did not meet the requirements of the company regarding accuracy.

Southwest Airlines reduced costs on fuel at the expense of Big Data
Southwest Airlines reduced costs on fuel at the expense of Big Data

In a pilot project the Alteryx Designer platform was used. It helped to construct 8 different forecast models which included function of modeling of regression of a time series and neural networks. For every month and each airport a system could generate 9.600 forecasts. Data processing time for drawing up forecasts managed to be reduced by 60%. Besides, forecast accuracy was increased. The project helped to find out Southwest Airlines that acquisition of fuel at one supplier will provide big benefit, than at several as it practiced earlier. Speed of drawing up forecasts managed to be reduced from 3 days to 5 minutes.

The Danish airline Thomas Cook Airlines Scandinavia rendering services of charter transportations selected the solution GoDirect Fuel Efficiency of development of Honeywell for increase in efficiency of use of fuel and reduction of harmful emissions[11] in May, 2017]. Software provides meeting requirements of the EU for regulations of emissions of harmful substances.

Thomas Cook Airlines Scandinavia uses developments of Honeywell for increase in efficiency of use of fuel and reduction of harmful emissions
Thomas Cook Airlines Scandinavia uses developments of Honeywell for increase in efficiency of use of fuel and reduction of harmful emissions

The complete solution GoDirect including a number of services and applications provides operators, crews of vessel and repair teams with the software for management of large volumes of data and analytical tools for increase in efficiency of the airplane. The efficiency of application of the given solution for economy of fuel differs depending on models of airplanes and airlines, however the existing clients tell up to 5% about annual economy. For large airlines with a big vessel fleet it corresponds to saving tens of millions of dollars annually.

For February, 2019 software of GoDirect Fuel Efficiency is already used on all vessels of the Thomas Cook group.

Forecasting of delays of runs using Big Data

In April, 2018 the venture division of JetBlue Airlines, the American low-cost airline, invested means in Lumo startup (the last name FlightSayer)[12]. According to the management, a startup, the company is capable to predict delays of runs in several hours and even several days to a departure.

JetBlue Airlines predicts delays of runs using Big Data
JetBlue Airlines predicts delays of runs using Big Data

The solution of the company is under construction on technologies in the field of artificial intelligence. Users can check the accuracy of work of the solution Lumo free of charge in the online mode. For this purpose they should enter on the startup website date of flight, the flight number and the name of airline. After that users will receive the forecast with indication of duration of a possible delay. Use of the solution Lumo allows airlines and travel agencies to rebook tickets of passengers, saving that their loyalty and also preventing the crisis situations at the airports arising at delays of departures.

Among possible problems when implementing the solution experts select, first of all, existence in the market of the competing solutions of large players and also insufficient integration with the systems of travel agencies. As for the competing solutions, experts call SITA FlightPredictor (in April, 2017 the management of SITA announced emergence of the starting version of the solution by the end of summer of 2017); offered about 2016 so-called Schedule Recovery System from Amadeus which first user was Qantas; the program of Google (since November, 2017, it is expected passengers). A number of startups also offers similar programs, for example, of Freebird from Cambridge.

At the solution presentation the startup specified that by spring of 2018 the pilot with large international airline should be implemented. However, any additional information on this subject it was not published.

Application of Big Data for the analysis and demand forecasting on air transportation

Southeast Asia can be considered as rather strong player in the field of predictive analytics of demand for air transportation.

So, in December, 2016 the Philippine Airline selected the solution of PROS company for optimization of global strategy in revenue management (Global Revenue Strategies)[13]. The cloud solution of PROS Origin & Destination allows to predict demand and to optimize income taking into account market trends, the shifts in demand and the pro-active analysis.

Philippine Airline optimizes income taking into account market trends, the shifts in demand and the pro-active analysis
Philippine Airline optimizes income taking into account market trends, the shifts in demand and the pro-active analysis

Philippine Airline is national airline on Philippines. Airplanes of the company for February, 2019 fly in 30 local and 43 international directions. The company operates 79 aircrafts ― Boeing 777-300ER, Airbus A340, A330, A321, Bombardier Q400 and Q300.

The developer in the field of technologies for aircraft of Sabre Corporation, the developer of solutions for world tourism industry, announced in November, 2017 signing of the long-term agreement according to which the Hong Kong Airlines will receive Big Data-solution MIDT (Market Information Data Tapes)[14]. This solution represents the database providing access to historical and forecast (depth up to 11 months) to data on armorings. Possession of this product allows airlines to analyze influence from measures in the field of formation of rates, marketing programs.

Hong Kong Airlines uses Big Data-solution MIDT for implementation of marketing strategy
Hong Kong Airlines uses Big Data-solution MIDT for implementation of marketing strategy

Hong Kong Airlines expects to use a product when implementing plans at first of work in the market of North America. The product allows to create reports and has analytical functionality that gives to the user the chance to reveal optimal channels for implementation of marketing strategy. Hong Kong Airlines will get data access of agents of Sabre worldwide at all points of a departure and appointment where there is an airline.

Big Data for increase in level of customer satisfaction and personalisation

The British Airways company entering TOP 10 on a passenger traffic uses since 2013 Big Data Analytics for increase in the service level of the clients: The carrier collects different data on passengers in special storage, and then loads them into the program under the name "Know Me"[15]. ― to learn the program purpose and better to understand customer needs and also to use the data which are saved up during various contacts with these clients for increase in level of their service.

British Airwaysispolzuyet Big Data Analytics for increase in the service level of the clients
British Airwaysispolzuyet Big Data Analytics for increase in the service level of the clients

"Know Me" contains different data on passengers: behavior at online orders, wishes when shopping, preferences when choosing the place. All this information is automatically generated and automatically used at the following armoring performed by the client.

The program works with the help of the analytical software from Opera Solutions. Also search in a photo of Google Image Search is used that allows the staff of airline to distinguish especially important and a lot of the flying passengers already while they are included into the airport or a business box, and, respectively, to offer them service of the top class.

Other large player of the market, Virgin Australia at the end of 2017, reported that it is engaged in optimization of work of the applications in the field of machine learning[16]. The company attracted DataRobot startup to these purposes. The American startup developed the platform of predictive analytics for fast creation and implementation of forecast models. This platform already helps Virgin Australia to reduce time for creation of forecast models for 90%, forecasting accuracy increases at the same time by 15%.

Virgin Australia works on optimization of the loyalty program of Velocity Frequent Flyer, implementing in it predictive analytics
Virgin Australia works on optimization of the loyalty program of Velocity Frequent Flyer, implementing in it predictive analytics

For February, 2019 the airline works on optimization of the loyalty program of Velocity Frequent Flyer, implementing in it predictive analytics which should support clients of the company when choosing of the best time by them for use of the received balls. The task of creation of forecasts/models of the one who with the maximum probability is ready to go to a travel what price and what type of a travel is preferred by the traveler is set for DataRobot. In general, it is about increase in level of service for participants of the loyalty program of airline.

Use of smart chat-bots

In September, 2017 the Finnish national airline Finnair started the chat-bot based on artificial intelligence for the account in Facebook Messenger Finnair launches messenger chatbot[17]. The chat-bot Finn is capable to sell tickets, to provide information on time of departures and to what quantity of baggage can be taken aboard. Besides, the bot can direct passengers to the page of Manage My Booking airline where they can purchase additional services. Also the bot is capable to answer most frequently asked questions. In case of impossibility to give the answer, it redirects the passenger to employees of a client service.

Finnair was started by the chat-bot based on artificial intelligence
Finnair was started by the chat-bot based on artificial intelligence

Finn understands and communicates in the English and Finnish languages. The company considers the possibilities of use of a bot on other platforms, for example, on Wechat in China. The chat-bot was created together with Caravelo company which is engaged in development of solutions for the aviation industry.

The chat-bot based on artificial intelligence was let also into Air New Zealand airlines under the name of Bravo Oscar Tango (Oscar) on the website of airline in New Zealand. In October, 2017 it entered foreign market – the Air New Zealand's AI chatbot Oscar says g'day to Australia market[18]. The task of the chat-bot consists in the answer to FAQ-questions of visitors of the website.

The intellectual chat-bot uses also Air New Zealand
The intellectual chat-bot uses also Air New Zealand

The started beta of a bot became initial development of Air New Zealand in the field of artificial intelligence. Since launch it studied based on questions of visitors of the website. According to the technology head of airline, in the first day of work the chat-bot could answer 7% of requests successfully. As of October, 2017 this indicator reached 67%, and by February, 2018 ― 75%, at the same time the chat-bot continues learning process. As found out airlines, passengers, mainly, address the chat-bot when prompt replies to the questions arising in day of a departure or concerning armoring of tickets are required for them. When armoring tickets such subjects as obtaining the booking confirmation, admitted weight of baggage and an aviamila are the most popular. In general, the chat-bot is capable to conduct conversations on more than 380 different subjects. For August-October, 2018 the chat-bot led about 55.000 discussions. On average in day he received 300-350 requests. In some days the number of requests exceeded 1.000.

As representatives of Air New Zealand said, before start in Australia the chat-bot learned some expressions from the Australian slang and a number of sayings. Along with the answer to questions the chat-bot can sing and tell jokes. Recently it was started in mobile application of Air New Zealand.

Other Australian airline, budget Jetstar started in February, 2018 the virtual assistant to "Jess" on[19]. Initially the chat-bot began to work at the web page of airline in 2013. In November, 2017 testing of the solution for Facebook began. Jetstar became the first airline in the Pacific Rim which expanded work of the chat-bot from the web page on Facebook Messenger.

Jetstar started the virtual assistant to \"Jess\"
Jetstar started the virtual assistant to "Jess"

"Jess" answers questions of clients concerning their armorings, provides information on baggage and places. It is available to clients from Australia, New Zealand and Asia.

The solution uses Natural Language Understanding technology from Nuance Enterprise company, technologies of expanded permission that allows it to conduct dialogues in real time. However the chat-bot is not capable to initiate conversations or to do offers.

The airline announced results of work of the chat-bot on the platform: The effectiveness of its work (a share of questions on which the chat-bot could give the answers which satisfied clients) was 73%. Time of the answer to a number of questions decreased from 17 to 0 o'clock minutes, the head of a client service of Jetstar reported. As for the general indicators of work of the chat-bot on two platforms, monthly it is involved in 250.000 dialogs, and all since launch he participated in 9 million dialogs with clients of Jetstar.

In April, 2018 on Passenger Experience Conference the English company Aviget showed the chat-bot with advanced functionality[20]. Using machine learning technology, the solution allows to interact with passengers based on different platforms, such as Facebook Messenger, Viber and WeChat. The solution is used by Natural Language Processing for recognition enough difficult complex phrases, for example, of "find me a flight from Heathrow for under of $150 that arrives in Jakarta on Boxing Day".

Ensuring aviation security using artificial intelligence and Big Data

In 2018 the British airline Virgin Atlantic reported what it is going to implement in the program runtime of the British[21]. It is about the Flight Risk Assessment System system which uses artificial intelligence technologies (AI) and machine learning (Machine Learning) for processing of data bulks for the purpose of increase in efficiency and reliability of aviation transactions.

Virgin Atlantic applies artificial intelligence technology and machine learning to processing of data bulks for the purpose of increase in efficiency and reliability of aviation transactions
Virgin Atlantic applies artificial intelligence technology and machine learning to processing of data bulks for the purpose of increase in efficiency and reliability of aviation transactions

A system collects data from more than 200.000 sources in 60 different languages. Then these data are transferred to a DB which contains information on 380.000 events in the field of aviation security and reliability. Data are collected from open sources, such as messages in media, from social networks and from industry websites. The Osprey system provides quick response to threats and long-term data analysis for identification and knocking over of the reasons of these threats. The system purpose ― to provide participants of the aviation industry with information, necessary for acceptance of the right decisions, regarding security of flights. Thanks to an opportunity to analyze historical events, using machine learning and artificial intelligence, a system is capable to predict the probability of, where and when they can repeat.

The Osprey system divides the globe into the blocks which are not limited to borders of the countries or borders of Flight Information Region. These blocks are dynamically updated. Other data sets, such as height of the area and population density also can be added to information on runs in these blocks. Thanks to it it is possible to define the existing or forming trends.

The analysis and assessment of marketing initiatives by means of the solutions Big Data

In May, 2018 developer of the analytical solutions Neustar, Inc. reported that the Scandinavian Airlines selected its product for assessment and measurement of influence of marketing for sales[22]. The solution under the name Neustar MarketShare will provide airline with analytics which will allow to plan daily marketing activity and will give support at adoption of marketing solutions.

SAS (Scandinavian Airlines) selected the product Neustar for assessment and measurement of influence of marketing for sales
SAS (Scandinavian Airlines) selected the product Neustar for assessment and measurement of influence of marketing for sales

The approach "Marketing Mix Model" (Marketing Mix Modeling, MMM) implemented by Neustar company in its solutions gives the chance to airlines to analyze key sales channels and the main regions for the choice of optimal model of implementation of the products. Besides, to give to Scandinavian Airlines a complete idea of their activity in mass media and influence of different drivers on economy of the company, the solution Neustar MarketShare analyzes different resources – the media which are not relating resources, data on economy, the competition, seasonality to media.

Big Data and other spheres of civil aviation

Big Data technologies find the application and in other spheres of civil aviation.

Spafax developers, provider of entertaining solutions for airlines, created in June, 2018 basic working prototypes of solutions based on artificial intelligence for their integration into the onboard entertaining platform with functionality of personalisation which is used by a number of airlines, including American Airlines, Lufthansa and[23].

The first solution ― model of the chat-bot which is approximate to human communication. At the same time for improvement of dialogue opportunities the application based on machine learning under the name LUIS (Language Understanding Intelligence Service) was used. Besides, cognitive services, in particular face recognition are integrated into the chat-bot. Thanks to it clients of airlines will be able to request for viewing onboard the list of movies in which a certain actor plays. For this purpose it is required to load into the application the photo of this actor only.

The second solution ― the application based on artificial intelligence for the analysis of video content using machine learning. The platform had an opportunity to reveal certain objects, scenarios or content with age limits that often is required according to requirements of airlines for content. For example, the artificial intelligence is capable to detect content connected with scenes of the crash of airplanes or the adult contents and to filter it.

In April, 2018 the FoxTripper company for the first time showed the program with "the moving card"[24]. The program provides to passengers information on those places over which the airplane flies by, and allows passengers to perform armorings in destination points. The data collected in flight in a combination with data of airline on the passenger allow to build forecasts relatively total what products and services are for it relevant.

Other interesting example ― Gogo Air. This developer company of an information and entertaining system for passengers in flight uses artificial intelligence and machine learning to help airlines to increase the level of the rendered services[25]. Gogo Air uses tool kit of the Adobe Analytics series, including the virtual analyst (Virtual Analyst) ― the tool based on machine learning, for collection of information on clients for a number of large airlines.

Providing entertaining content and Wi - Fi - access in flight, Gogo Air collects information on the passengers using these services. Then this information is exposed to processing and the analysis. As a result of airline obtain those data which help them to improve customer service and, often, to offer the passengers more targeted products. Airlines learn in what products their clients can be interested during flight what devices they use in flight, what is the time they are ready to see off on the Internet or what entertainments they prefer in the airplane.

Airlines use the obtained data for personalisation of services based on a situational context, for example, adapting for the client screens of the information and entertaining systems in the airplane depending on length of flight used by the passenger of devices, a destination point.

Also technologies of catering services onboard do not stand aside. So, in April, 2018 in Hamburg the Black Swan Data company developing solutions for data analysis signed the agreement on cooperation with[26]. The cooperation purpose ― data analysis of passengers and trends on social networks for forecasting of what menu from the airplane will be selected by passengers. Passengers will be able to order and expect receiving the favourite dishes after landing. The pilot project of two companies showed quite good results: It was succeeded to reduce waste on food by 50% and to increase performance by 15%.

In May, 2018 the company ― the developer in the field of solutions for aircraft of SITA offered the system of tracking and management of baggage. The BagJourney technology developed by it allows to manage transactions with baggage increasing to number[27]. Only for the first six months 2018 more than 20 carriers selected this solution. SITA BagJourney ― one of the main solutions which helps the aviation industry to execute resolution 753 IATA in which the requirement for tracking of baggage at each stage of a travel is stated.

The solution SITA BagJourney is used every year for processing of hundreds of millions of articles of luggage. According to users, the solution lowers number of errors by 30%. BagJourney is compatible to different hardware, including mobile devices for scanning or fixed instruments.

According to BahamasAir, one of users of the solution, after its implementation within 7 days process of complete transition to mobile devices for tracking of all baggage in two directions which are the most loaded in respect of baggage ― was succeeded to perform Nassau and Miami. Based on six months the number of complaints to problems with baggage on the most loaded direction decreased by 60%. The airline is going to implement the solution on all directions and calculates that until the end of the year it will conform completely to requirements of resolution 753.

Interview with experts

Kirill Bogdanov, CIO of Aeroflot told about opportunities of technologies of the analysis of Big Data for an aviation industry and the project of creation of the platform for the processing of customer appeals constructed using Big Data in an interview of TAdviser.

Dmitry Bulenkov, the vice president for sales of RAMAKS Group, in an interview of TAdviser told about use of Big Data technologies in the civil aviation and projects executed by the company in this sphere.

Use of Big Data technologies in the Russian civil aviation

For February, 2019 Big Data is already actively used in the Russian airlines. Aeroflot, S7 and other large players of the market work with Big Data. The first developments in the field of Big Data appeared in a domestic segment in the middle of zero, they are connected with creation of services of customer service, armoring of tickets and repair of aircrafts.

Big Data technologies find application in the Russian airlines
Big Data technologies find application in the Russian airlines

In this chapter we will dwell upon the overview of the market of domestic developers and system integrators in the field of civil aviation, we will select projects on the basis of Big Data and we will tell about the most remarkable of them. The classical vendors working in an aviation segment ― with Sabre, Lufthansa Systems, SITA, INFORM, Amadeus. These companies develop specialized solutions for the industry.

At the same time in the field of the developments connected with Big Data also such vendors as, for example, IBM and Oracle work. The Russian vendors working with Big Data in a segment of aircraft – Group with the Tarantool platform, Innodata company with Kribrum technologies[28] and Hadoop Apache Hadoop[29].

Also in Russia it is provided many system integrators which are engaged in implementations in the aviation industry. Among them:

  • RAMAKS group (25 years in the market, the technology consortium of system integrators and developer companies covering all range of needs of the customer — from development of strategy before support of complete solutions);
  • Integro Technologies (dynamically developing system integrator providing services in the development area, complex implementation and support of specialized IT solutions and also software localization. The company offers a wide choice of complete solutions for implementation of large-scale projects for the transport sector, in particular for airlines, the airports and land services. Enters into RAMAKS Group);
  • Technoserv Group;
  • Gazpromneft Avia;
  • SAP CIS.

The Russian airlines "Aeroflot", Russia and S7 actively implement Big Data technologies. Further we will concern the most perspective and innovation projects based on the specified technology.

Automated system of maintenance of the flight validity using Big Data technologies

In 2017 the automated system of maintenance of the flight validity, maintenance and repair of aircrafts on the AMOS platform from the Swiss vendor of AMOS Swiss Aviation was put into commercial operation in Aeroflot. Thanks to this project tracking of a real picture of a status of the air transport became possible. Consolidation and transformation of the sources given from huge number, including unformalized data is made; integration of the systems of maintaining a warehouse stock is implemented; involvement of a set of structural divisions is provided; increased requirements to quality of transfers for project needs are met.

In Aeroflot the automated system of maintenance of the flight validity, maintenance and repair of aircrafts is introduced
In Aeroflot the automated system of maintenance of the flight validity, maintenance and repair of aircrafts is introduced

Use of AMOS at the expense of such indicators as performance improvement of work of staff of the relevant divisions reduction of warehouse stocks and process optimization of planning and material security of a MRO significantly cut down expenses on maintenance of AF for 4%. Increase in efficiency of planning and accounting of Inventories for 4% thanks to integration of the AMOS system with the production SAP systems and to high quality of data in them.

It is also necessary to note decrease in labor costs during the work with a system and ensuring automatic formation of working cards for service of the CU due to loading of plane documentation, absolutely other landscape and the organization of a reliable restoration system of data allowed to increase significantly fault tolerance of the solution, to increase quality and efficiency of formation of the accounting and management reporting for providing to regulating authorities in the field of aircraft.

This project allowed us to have near at hand a real picture of a status of the flight validity and maintenance and repair of aircrafts of Aeroflot. Using the AMOS system we received a global picture of all processes on a MRO in the single integrated system: purchase planning, planning and accomplishment of repairs, online integration with the Boeing and Airbus systems. It allowed to reduce labor input, to increase quality of processes of a MRO and maintenance of the flight validity of aircrafts. Our next purpose — to reduce a downtime of the aircraft on the earth thanks to transition to predictive maintenance. We will start implementation of this task in 2018.
Kirill Bogdanov, CIO of PJSC Aeroflot[30]

The platform for the analysis and processing of Big Data from social networks

In 2017 the RAMAKS Group on behalf of Integro Technologies implemented in PJSC Aeroflot the platform for the analysis and processing of addresses of passengers on social networks using platforms IBM BigInsights and IBM PureData.

Integro Technologies implemented in Aeroflot the platform for processing of addresses of passengers in social networks
Integro Technologies implemented in Aeroflot the platform for processing of addresses of passengers in social networks

Work with client reputation is of great importance for transport companies, including, and for aircraft. Social networks allow to collect in real time responses of passengers and to react quickly to them.

Advantages of the considered system ― a possibility of continuous tracking of satisfaction with the company and interactions with users in social networks; security and identification of the terrorist organizations, extremism and other problems; permanent improvement of the offer for the customer by means of the analysis of Big Data in social networks and an opportunity to communicate with the operator directly; maintenance of reputation of airline by operational contact with audience in social networks; analysis of the user preferences and drawing up individual product offerings and also successful targeted advertizing. In more detail about the project - here.

Basis of all current and future future changes when approaching to work with clients are technologies of work with Big Data. Thanks to Big Data PJSC Aeroflot has additional opportunities of conducting effective business.
Kirill Bogdanov, CIO of PJSC Aeroflot[31]

Domestic Tarantool DBMS in the project of analytics of Big Data

Aeroflot implemented algorithms of predictive analytics of Big Data within the draft of the platform for the analysis and processing of addresses of passengers on social networks. As DBMS domestic development ― the solution Tarantool of Mail.Ru Group company was used.

Aeroflot used Tarantool DBMS in the project of analytics of Big Data
Aeroflot used Tarantool DBMS in the project of analytics of Big Data

The complex consists of a large number of modules which cover both functional business requirements, and modules of integration into the existing IT infrastructure of PJSC Aeroflot and different channels of receipt of addresses (social networks, e-mail, the official site, a personal account).

The first module is responsible for identification of the client on the basis of complex data analysis, both the text, and data of a profile of the author. The number of addresses can reach several thousand a day.

The second module is intended for search of doubles of addresses. Text copying for placement on different resources or sendings by mail. There is determination of semantic similar posts for the purpose of identification of the clusters which are incidents. Processing of several similar posts leads at once to significant reduction of loading of responsibles.

The third module "Newsmakers" is one of the most important in a system. Its main feature ― prediktivno to reveal dangerous posts still before growth of activity begins. Thus, the put algorithms indicate potential "info bomb" and give the chance to level reputation losses.

The settlement number of addressing data equaled to several thousand requests per second with a necessary response in couple of milliseconds. For the satisfaction of high requirements of the customer as, for example, the registered time constraint in three seconds on enrichment of the address different properties, was required use of the hi-tech software. Based on the carried-out tests on performance, quality of data storage and functionality, it was decided to use domestic development ― Tarantool DBMS.

Tarantool is used in the Platform as the operational database in which addresses are stored in a type of the special data structures necessary for analytics algorithms. Extremely high performance and existence in base of such properties as secondary indexes and support big the number of connections without performance penalty, allowed to implement successfully the above described functional modules, without going beyond the delivered time frames.

Use of domestic developments in such large company as Aeroflot, extremely important. The Russian software often does not concede in anything, and as in our case, and exceeds foreign analogs. For this reason Tarantool was selected. And, naturally, the important factor of import substitution is executed that for our company is one of key priorities for the next years.
Kirill Bogdanov, CIO of PJSC Aeroflot

Based on implementation the customer using the Platform considerably increased efficiency of processing of complaints and customer appeals by responsibles of PJSC Aeroflot, cardinally reduced delivery time of the address and time for the processing/solution of a question at the expense of mechanisms of enrichment of the address a context, tonality, subjects (tagging), the author's profile, etc. All this is aimed at achievement of positive economic and reputation effect practically at all stages of provision of services of PJSC Aeroflot. On the basis of successful experience all project participants will continue to use Tarantool software in the projects and to strengthen partnership[32][33].

The analysis of Big Data for customer segmentation

In 2018 the Technoserv system integrator implemented the project on creation of a system of intellectual customer segmentation for Aeroflot airline on the IBM platform. A system, using the analysis of Big Data and model of machine learning, carries out customer segmentation on a set of characteristics. IT solution helped airline to implement marketing strategy, to actively develop online sales and also to increase the number of passengers.

Technoserv implemented the system of intellectual customer segmentation for Aeroflot
Technoserv implemented the system of intellectual customer segmentation for Aeroflot

Technoserv confirmed that Big Data technologies in general are demanded in the transport industry, and as confirmation to it serves increase in number of projects with use of the specified technologies. At the same time the subject of projects, according to her, is absolutely different. It and problems of increase in personalisation of communications with clients, pro-active repair of the equipment, prediction of demand and other tasks solved using algorithms of machine learning and the analysis of large volumes of the structured, unstructured and semistructured data[34].

Carrying out calculations on a blockchain

In July, 2017 Alfa-Bank started a blockchain project based on the Ethereum platform for carrying out calculations between S7 and its airline by sales agents of tickets. At the same time Alfa-Bank acted as clearing bank. By such interaction it was succeeded to solve a mistrust problem in case of scaling of a system and connection to this platform of other banks or airlines. Implementation of the platform opened possibilities of essential business process optimization both for airline, and for her partners. Speed of calculations grew from 14 days to 23 seconds.

Alfa-Bank started a blockchain project based on the Ethereum platform for carrying out calculations between S7 and its airline by sales agents of tickets
Alfa-Bank started a blockchain project based on the Ethereum platform for carrying out calculations between S7 and its airline by sales agents of tickets
We carried out the transaction on purchase of the air ticket through an open blockchain of api to bank, but I am sure that such scheme will be used by many companies around the world soon. The blockchain platform allows to optimize business processes significantly. It automates any scheme of settlement, even very difficult — for example, warehouse deliveries. At such mechanism participation of the person is practically not required: it is not necessary to make out bills, to carry out reconciliations, to write acts. Potentially suppliers of an onboard power supply, fuel, airport services — all those companies can be connected to the platform with whom S7 Airlines constantly works and not only.
Pavel Voronin, the deputy CEO for information technologies of S7 Group[35]

Aviagas station on a blockchain

In August, 2018 Gazpromneft-Aero, the operator of aviafuel business of Gazprom Neft, and S7 Airlines developed and implemented the joint smart contracts (Aviation fuel smart contracts, AFSC) based on a blockchain. The project allowed to automate planning and accounting of supply of fuel and is designed to increase the speed of settlement when refueling airplanes.

Gazpromneft-Aero and S7 Airlines implemented the joint smart contracts based on a blockchain.
Gazpromneft-Aero and S7 Airlines implemented the joint smart contracts based on a blockchain.

According to the statement of representatives of Gazprom Neft, it is experience of use of technologies, the first for the Russian aviation market, of the distributed registers. With their help the airline had an opportunity to instantly pay fuel directly when filling in airplanes without prepayment, bank guarantees and financial risks for participants of the transaction. Such approach increases efficiency of financial transactions and reduces labor costs, consider in the oil and gas company.

Forecasting of breakdowns of the S7 Airlines airplanes using machine learning and the analysis of Big Data

At the beginning of March, 2018 S7 Airlines developed the system of predictive maintenance (predictive maintenance). According to the company, it became the first Russian airline which completed development of a similar system.

S7 Airlines developed the system of predictive maintenance
S7 Airlines developed the system of predictive maintenance

At the initial stage it is used for the Airbus A319 aircrafts. Further a system will be adapted for all plane fleet.

The system of predictive maintenance assumes the analysis of an array of historical data on maintenance of airplanes and work of separate components.

The software for data analysis and creation of a mathematical model was developed by specialists of S7 Airlines together with the Russian company Datadvance specializing in development of solutions for predictive analytics.

In March, 2018 the data array from 2012 for 2017 was already available to the analysis. These are the data recorded in the systems of telemetry of airplanes, databases of holding of maintenance and repair of S7 Technics aviation equipment and meteorological data.

The main objectives which the company expects to solve using predictive maintenance, - reduction of number of the departures delayed for technical reasons, increase in flight safety and efficiency of maintenance of courts, forecasting of probability of possible breakdowns for each airplane in the company park.

Equipment of airplanes of Pobeda airline RFID tags

Pobeda airline implemented the first-ever project of equipment of airplanes RFID tags within which on all crash equipment of each board the radio tags which are read out by means of the tablet computer are established.

The Pobeda airline equipped with RFID tags stock of airplanes
The Pobeda airline equipped with RFID tags stock of airplanes

Several hundred RFID tags in each of airplanes fasten literally to everything that is not nailed — from life jackets to seat belts. Also tags fasten on heat resisting gloves, megaphones, oxygen cylinders, masks, fire extinguishers, etc.

Project objective - to accelerate inventory of crash equipment which happens after each flight. One of stewards starts a special application on the tablet and passes on salon, scanning RFID tags. Each detected tag responds a short sound signal, and at the end the application generates the report on existence of all crash equipment. The report is right there loaded on the server: SIM cards are installed in tablets, and a cloud part is implemented based on Microsoft Azure.

If there is no equipment, it at once is visible in the report, respectively, in this case the command for departure of platform buses with passengers is not given and there is their check.

Without the equipment the airplane cannot be allowed to the following run (i.e. if onboard there is not enough life jacket — means, to one of passengers it will be refused transportation). Manual inventory takes away much more time and forces: vests alone under chairs — 189 pieces, and everything it is necessary to check them. Thus, thanks to RFID technology Victory managed to reduce the minimum turn-around time for the aircraft from 30 to 25 minutes. It is one of key KPI in passenger aircraft: the sense is that the less time goes from an arrival to the airport to a departure the following run, the cost efficiency of airline as the airplane brings income only is higher when it flies, and is not necessary on the earth. At the sizes of the fleet of Victory in fifteen airplanes reduction of time of inventory of each board for 5 minutes gives the chance to perform at least one additional flight, without increasing the park of aircrafts.

Creation of the center of innovations in the field of civil aviation for strengthening of examination in the field of Big Data

In 2017. The Innodata company, the Russian software developer in the field of the innovative technologies, and the Russian IT university "Innopolis" created the Center of Innovations in Civil Aviation (TsIGA). The consolidation purpose ― development of technology and digital presence at modern civil aviation, contribution to disclosure of an essence and value of the modern technologies influencing demand and supply for players of an aviation industry, integration of innovations of the digital world into the current technologies of civil aviation. In 2018 RAMAKS Group for the purpose of strengthening of the existing examination in the field of Big Data technologies, and to development of specialized solutions for the aviation industry became the partner of the Center.

Innodat and the Innopolis IT university created the center of innovations in the field of civil aviation for strengthening of examination in the field of Big Data
Innodat and the Innopolis IT university created the center of innovations in the field of civil aviation for strengthening of examination in the field of Big Data

The main activities are a development existing and creation of new solutions for an aviation industry, respectively. The center project implementation in the scientific and technical, innovation or information and analytical plane conducts both educational activity, and project, whether it be. TsIGA is also open also for pilot projects for the purpose of promotion of advanced technologies and solutions and is ready to give support in development.

The basis of the economy of the city of Innopolis ― the hi-tech industries therefore active participation in similar consolidation is important for us, one of problems of which - the report of value of modern technologies and their active implementation in an aviation industry.
Iskander Bariyev, the vice rector ― the head of department on project and research activity of ANO VO Innopolis University

The center conducts work in the field of the solution of such tasks as development and deployment of technologies of augmented and virtual reality for fight against an aero phobia, navigation at the airport on the basis of technologies of virtual reality, the behavioural analysis of activities of employees in an information field, prediction of purchasing power of passengers and formation of dynamic recommendations about a change in value of tickets, planning of the schedule of flights and the analysis of optimization of the seasonal schedule, predictive management of a passenger traffic, personnel management at the airports, development of the system of selection of personal package service offerings of airline and partners and also a technique of scanning of a surface of aircrafts during postflight maintenance, the analysis of a runway, layer management of overbooking, the analysis of interests of passengers and formation of offers for them.[36][37][38]


The examples reviewed above show that airlines ― already not just aircrafts, carriers to which we managed to get used. An important basis of their development ― technologies of Big Data which do possible, for example, personalisation of services. Individual offers which do a trip of each passenger of the most comfortable. Information search about a travel, booking, search queries – any actions leave digital marks which can be analyzed for formation of the most point service package in network. Besides, work with Big Data allows to raise a customer loyalty, for example, due to prompt reply to addresses of passengers.

Even more data generate production systems. Airplanes, railway locomotives and trains are a source of a huge flow of technical data which arrive from sensors, installed in engines and life support systems. The detailed analysis of these data allows to reveal and predict need of repair of this or that spare part. Thus, Big Data allow to raise the security level and also to save considerable means for carriers. Necessary time for repair is reduced and the airplane can be used directly during longer term.

The offered material mentioned some of opportunities and practical results of use of Big Data technologies in the aviation industry, in reality of such opportunities for development every day becomes more and more.


  1. of Insight from flightglobal: Big Data, big picture
  2. [ The Year Airlines Seriously Start Investing in Connected Aerospace Technologies
  3. Evaluating the Economic Benefits of Connected Airline Operations
  4. [ of Commercial aviation at forefront of innovation in artificial intelligence, digital twins, mobile applications, and unmanned aircraft
  5. of 35.2% of During 2016 to 2021 - ReportsnReports
  6. of EasyJet Talks Evolving Predictive Maintenance Operations at AEE
  7. to Delta's Maintenance Prognostics Will Continue On Newest Aircraft
  8. airline Honeywell und Cathay Pacific tests zu Big Data
  9. IBM’s Watson puts the AI in air travel
  10. of How Southwest Airlines Chooses Big Impact Analytics Projects
  11. [ of Thomas Cook Airlines Scandinavia Chooses Honeywell Software To Improve Fuel Efficiency
  12. of JetBlue's Venture Arm Invests in a Startup That Predicts Flight Delays
  13. of Philippine Airline Selects PROS to Optimize Global Revenue Strategies
  14. Hong Kong Airlines Purchases Sabre MIDT Network Plus Data
  15. Big Data At British Airways
  16. of How machine learning is helping Virgin boost its frequent flyer business
  17. [1]
  18. of Australia
  19. the Facebook Messenger Jetstar launches Facebook Messenger chatbot "Jess" platform
  20. of Aviget
  21. company Osprey Osprey to provide Virgin Atlantic with AI security and operations solution
  22. of Scandinavian Airlines Brings on Neustar to Measure Marketing's Impact on Key Business Drivers
  23. SWISS Spafax hackathon finds new airline use-cases for AI
  24. to Why airlines are finally poised to unlock Big Data to enhance the passenger experience
  25. of How data analytics enhances the experience of air travel
  26. gategroup Why airlines are finally poised to unlock Big Data to enhance the passenger experience
  27. of SITA's BagJourney Assisting Industry with Bag Tracking airlines
  28. Kribrum
  29. [2]
  30. Implementation of an automated system of maintenance of the flight validity and maintenance and repair of aircrafts
  31. of#best decision/42/project/958 the Platform on work with customer appeals
  32. Aeroflot implemented Tarantool DBMS for work with Big Data
  33. Aeroflot implemented the Russian Tarantool DBMS
  34. Technoserv segmented clients of Aeroflot
  35. Alfa-Bank and S7 Airlines started sale of tickets through a blockchain
  36. Innodat and Innopolis will be engaged in innovations in civil aviation
  37. of "Ramaks" became the partner of the Center of innovations in civil aviation
  38. of TsIGA