By: Colonel Harpreet Singh Jaiswal (Retired)

According to the International Air Transport Association’s (IATA) Annual Safety Report 2023 Executive Summary, accidents have reduced by 61% over the last decade. Barring one fatal crash of 2023, the year enjoyed the lowest fatality risk rate of 0.03 per million route sectors. This translates to roughly one accident in about 8.8 lakh flights. This is due to aviation stakeholder’s willingness to, earnestly investigate accidents to improve safety while, undertaking aggressive research and development (R&D). However, every now and then, we still come across news of flight disruptions. These disruptions can be categorised primarily into three areas: Equipment or sub-assembly failure, turbulence related in-flight injuries and accidents and lastly, misadventures by unruly passengers.

Whatever be the cause of the disruption, it leads to unforeseen expenses in terms of urgent public relations management exercises, legal penalties, increased unscheduled maintenance costs and of course, the loss of goodwill. In addition, the airline’s public relations officers or corporate communicators often have to work in overdrive to salvage the airline’s reputation.

 Today, digitization which includes the Internet of Things (IoT), coupled with modern testing technologies, has advanced to such an extent that, it is possible to minimise the damage caused by the aforementioned factors. While these risks can never be completely eliminated, they can be minimised through digital transformation by the application of Artificial Intelligence (AI), rapid broadband digital communication and super fast computing for data analytics, amongst other tools.

The advancements in aircraft’s diagnostic through R&D, destructive and nondestructive material testing and a concentrated effort by all aviation stakeholders to find the root cause of any failure, have improved aviation safety and minimised losses caused by black swans.

The current era of digitization integrates artificial intelligence, the Internet of Things (IoT), machine learning, data analytics, fast computing and reliable signal transmission, to process inputs from diverse fields in order to improve operational safety and reduce human errors in civil aviation.

This has been made possible by the quick capture of real- time data from the sensors of airborne aircrafts, meteorological balloons carrying radiosondes, satellite imagery and wind or weather measurement stations. These sensors generate large volumes of data, known as big and fast data. This data is quickly collected, collated, segregated and processed using customised algorithms run by super computers that apply artificial intelligence and machine learning as operational resources to deduce actionable insights for reducing bottom lines.

Digitization also enables this raw and refined data to be sent in near real-time to aircrafts while they are in flight. These insights are also utilised in evaluating and optimising preventive maintenance schedules, managing turbulence complexes (which can lead to increased flying time due to flight path deviations), reducing fuel burn and carbon footprints.

 As far as managing unruly air passengers is concerned; a neural network analysis of social media data footprints generated by the public is used to identify potential troublemakers who might misbehave inflight. Such individuals increase aviation costs by necessitating the deployment of air marshals, cause flight disruptions, incurring legal and public relations or corporate communication expenses due to their misbehaviour.

The as far as the third factor mentioned in the above disrupters ie equipment failure; material failure rates have been minimised due to digitisation and systematic R&D. Consequently, injuries due to material failure are the least likely to occur in an aircraft today. Turbulence related flight disruptions and their associated costs are more plausible and profound.

 An aircraft consists of various assemblies and sub-assemblies. These components have undergone extensive testing, for shelf and operational life estimations. Consequently, the accurate Mean Time Between Failures (MTBF) articulation of components; enables aircraft maintenance repair and overhaul (MRO) engineers to schedule their maintenance plans in a pragmatic way. This automatically minimises the fixed costs incurred by aircrafts during maintenance repair and overhauls schedules.

The use of modern non-destructive testing techniques, such as atomic absorption rate analysis and modelling of product deterioration graphs based on factors like excessive aircraft vibrations, temperature extremities, stress and strain, have advanced considerably. Accordingly, data on component failures is interpreted using AI models, which can determine whether an extension to component life can be granted to an assembly or sub-assembly. This approach reduces the bottom line of aircraft operating companies by enabling them to make the most of the, remaining useful life of a component without prematurely discarding it; thereby increasing profitability.

This is where a significant business opportunity lies for Indian engineers and scientists. They can establish aviation sub-assembly test benches to evaluate the residual life of equipment, thereby reducing costs. This presents a never-ending stream of revenue generation and business opportunities, especially for the pre used component market.

Turbulence is another operating cost enhancer. It refers to atmospheric disruptions in terms of wind speed, direction, temperature, humidity, cloud cover, etc. Usually, it is the meteorological department that studies the current weather parameters, correlates them with past data and applies climate models to predict turbulence in a given spatial dimension in the foreseeable future. It’s worth noting that the first supercomputer Cray and India’s Param were used for meteorological predictions.

Today, with advancements in artificial intelligence and big data management generated through an increasing number of ground weather stations, aircraft capturing relevant atmospheric data while flying on a real time basis, satellite imagery and the use of radiosondes have improved turbulence prediction significantly.

The key point here is that all these sensors whether flying or stationary, are generating real-time, fast, and large amounts of data, which is instantly transmitted to ground stations using high-power broadband digital communication. Ground based computers run software to process this data and generate near-real-time turbulence predictions in three dimensions. This data is then beamed back to the operational control centres of airlines or directly to the pilots.

The pilots can use this information to make calculated decisions regarding potential deviations from the flight path to avoid turbulence and ensure passenger safety. It’s important to note that if the deviation from the flight itinerary due to turbulence is large, fuel consumption and flight time increases. This impacts the operational costs and profitability.

Because the predicted turbulence cuboid, created through real-time data analysis, shows varying intensity from its core to the fringes, flight engineers can make informed decisions as to how much deviation is necessary to balance safety with cost overruns. This critical decision cannot be left entirely to computers. It requires human discretion and experience who run quantitative tools like Maximising and Minimising equations to find optimal parameters. This is equation-based application is taught in B Com (H) and MBA curricula in India

In turbulent zones, an aircraft tends to buffet up and down. Consequently, flight attendants, who are often serving passengers at such times, may be at risk as also some passengers, who despite warnings, refuse to fasten their seat-belts. If the aircraft encounters significant turbulence, the aircraft buffeting may throw passengers off their seats. This causes injuries and trauma to revenue-generating passengers which may impact the airlines goodwill, safety record and bottom and top lines as enumerated subsequently.

The turbulence related flight path changes not only increase the operational costs of the flight, but the airline may also need to spend on compensation for stress management, medical expenses and passenger injuries. Such delays also force it’s public relations machinery into overdrive to salvage the image of an airline that has caused human injuries and experienced delays in landing slots. Such delays can affect the subsequent itineraries of passengers who want to take other flights.

Moreover, it is likely that an aircraft facing significant disruptions may not be assigned the scheduled and economically favoured landing slot and docking site at cargo terminals which are equipped with air bridges and vestibules for passenger disembarkation. This adds to unforeseen expenses for the airline.

Usually, turbulence data is provided by the International Civil Aviation Organization (ICAO) and other agencies to airlines. Turbulence estimation presents an economic opportunity for Indian companies by leveraging our IT industry, captured meteorological data, fast data communications and our extensive computing capacity to sell meteorological advisories for aircrafts operating within our airspace or even globally.

Entrepreneurial companies could also sell this data to visiting airlines operating in India and generate valuable foreign exchange. This could potentially become a significant revenue stream for the Indian economy as it strives to achieve the trillion dollar mark once these services are scaled up.

Human beings are generally well behaved. However, there are always some outliers to this statement. These individuals may include disruptive passengers who usurp other passengers’ legroom, engage in sexual harassment or inappropriate touching due to close proximity. Then there are fliers, who constantly get up to walk on the aisle and disturb others, or those who create food-related arguments. Additionally, issues such as liquor intolerance, air travel fears, or claustrophobia manifest disruptive behaviours in some passengers which is a nuisance.

In such cases, these individuals may occasionally harass fellow passengers or even engage in altercations with the airline crew or other passengers. This necessitates restraint by Air Marshals and even premature deviations from the planned itinerary to offload the disruptive passenger, involve the police and incur associated legal costs.

Moreover, in today’s digital world, it’s common for bystanders to record such incidents on their mobile phones and share with friends as soon as they land. In these scenarios, the airline’s corporate communication team perforce, goes into overdrive for damage control. This may include offering compensation to the aggrieved parties. All these factors ultimately increase the operational costs for the airline and eat into their net profits.

Such costs and vulnerabilities cannot be eliminated, but the use of social media analytics and neuroscience-based algorithms developed with psychologist’s insights gleaned from the social signatures left by potential passengers on the internet, can help in creating individual risk profiles of flyers. By analysing the social media activity of individuals, airlines can better understand and categorise potentially risky passengers.

This is not difficult because most people today love to maintain a digital presence across various platforms in audiovisual or textual modes. They often showcase their existence and activities on social media. This way individuals unknowingly disclose their personality traits, which can be analysed by trained professionals.

While the current level of data analytics and behaviour modelling is not highly advanced, still, outliers or trouble makers can be identified with a considerable degree of certainty. Airlines can take proactive measures, by designating certain individuals as banned flyers or charging higher ticket prices to cover the costs of additional security measures like deploying air marshals and other resources needed to manage potential disruptions caused by such passengers.

It is important to note that increasing ticket prices will not necessarily deter the disruptive behaviour, but some costs associated with disciplining and public relations efforts incurred to manage such people can be recouped through this process. Additionally, segregated aircraft seating arrangements can be structured to minimise the potential of flight disruptions caused by unruly passengers.

It’s a well-known fact that criminals are more likely to commit crimes if they believe that they can get away with it. However, if there is active surveillance and engagement by air marshals inflight and by the security staff at the airport, then; there is a possibility that a potentially disruptive individual may refrain from misbehaving. In such cases, the airline can not only ensure a smoother flight experience for others but also earn additional revenue from the disruptive outliers.

The idea of placing trouble creators on a non-flying list is a legitimate approach of managing passengers with dubious behavioural characteristics. This measure helps in reducing airport and flight disruptions caused by such individuals.

The digital era in aviation safety has indeed come of age. It offers solutions to reduce operating costs by minimising vulnerabilities. This includes creating a financial matrix-based comparison of turbulence related detours, estimating the residual life of aircraft assemblies, as well as predicting deviant human behaviour. These facets are then compared to the anticipated losses due to the activation of the aforementioned flight disrupting factors and economically or socially beneficial decisions are arrived at.

The implementation of the aforesaid of digital transformation ie capturing and utilising big data, integrating sensors, fast computing and reliable broadband communication enables risks to be minimised against unforeseen events that may impact an airline’s profitability. While these costs cannot be entirely eliminated, they can certainly be minimised.