Bitcoin’s Lightning Network

As Bitcoin’s popularity continues to rise, so too does its global count of transactions.

All blockchain technology is at its core a shared database that is “distributed” among the nodes, allowing anyone to see all the transactions recorded on the chain. Transactions that are recorded on a blockchain network are unsurprisingly called “on-chain” transactions.

This builds trust and security. However, Bitcoin suffers from an issue of scalability. It has a limitation to how many transactions it can process in one second and in one block.

The Lightning Network is a potential solution to this problem. We will discuss the Lightning Network, how it solves the scalability issue, a few issues the lightning network itself must overcome. 

What is the Lightning Network?

At its heart, the Lightning Network is a system allowing network participants to transfer their bitcoins between each other without any fees (or with minimal fees) by using their digital wallets. A payment channel is created between the two users (sender and receiver) so they can transact “off-chain” with each other. The Lightning Network is a layer on top of Bitcoin’s blockchain network to process smaller payments between participants. 

This system means that on-chain transactions will be fewer, and the total processing time will be reduced

Image courtesy of RAKESH SHARMA

With the Lightning Network, funds are transferred as quickly as the users’ wallets can communicate on the net. When business is concluded, there is a “closing transaction” on the main Bitcoin blockchain (layer 1) to settle all of the transactions.

The Bitcoin blockchain will not know how much each Lightning Network user owes until the bill is settled. Because transactions are generally not between trusted users when payment channels are opened, both sides place deposits of equal to higher values than the transactions themselves.

Is It Safe?

If, at any point, either user of a transaction wishes to back out, they can easily take their deposit and leave without consulting the other party. With such a one-sided withdrawal, the leaving side is required to wait for 1,000 block confirmations (approximately one week) to get their deposited bitcoins back. The party who did not leave will receive their security bitcoins instantly.

There are fraud measures built into Lightning. If one party tries to avoid paying the other, then the former suffers a penalty of forfeiting their whole deposit. 

The Lightning Network also allows users to jump through connected payment channels. These “network channels” allow indirect connections of payment channels through intermediaries. This is how the creation of unnecessary payment channels is avoided as scale increases.   

Ultimately, the lightning network works well for small transactions. Even if there are hypothetically over 1,000 transactions between users, the main blockchain shows only transactions: the first opening the payment channel and depositing money, and second closing it and settling the bill. All the transactions in between were feeless and instant.

Why is the Lightning Network Needed?

The general structure of Bitcoin’s network means that, when a transaction happens, it’s added to the newly created block once verified. The blockchain structure shows us that all its nodes each have a copy of the transactions for the vilification process.

In Bitcoin’s structure, these nodes are miners. Miners can only process a certain number of transactions per block, averaging 1,609, and there is only one block produced every 10 minutes. The network does lag down if the transaction volume becomes excessive. This leads us to Bitcoin’s scalability issue, where the network slow down significantly when it tries to process many transactions simultaneously. 

This also leads us to increased transaction fees, chipping away at a central, founding tenet of Bitcoin. And users can occasionally offer to pay a higher fee to have their transactions processed sooner, similar to traditional wire transfers. Small transactions will suffer greatly from these fees.

The purpose of the Lightning Network is to cut away these transaction fees through scale and enhanced processing capacity. For Bitcoin to compete competing against the likes of Ethereum or Tether, it must improve its scalability.

Problems with Bitcoin’s Lightning Network

However, there are a few issues which the Lightning Network still does not solve.

There Are Still Fees

While Bitcoin’s congestion is one of several factors influencing its transaction fees, the cryptocurrency’s own fee is still a significant component of the Lightning Network’s overall costs. 

There are also opening and closing fees which must be paid. The opening fee and its required deposit must be made on-chain. Once open, users process several transactions between each other or through network channels. But to settle a bill, the closing transaction is resolved on-chain as well. The deposited capital is tied up so long as the payment channel remains open. 

There is also a separate routing fee occurring when there are payments between payment channels via other network channels. These fees are low, and the idea is to have them be low enough to attract more uses, but they are still required. 

Let’s not forget the possible catch-22. If the fees via Lightning are too low, there may not be an incentive for nodes to facilitate these payments. As businesses adopt the Lightning Network, they may begin to charge fees for using it and negate its economic benefit. Some blockchains have already created solutions to this problem, allowing for cheap transactions through master nodes

The Lightning Network is Susceptible

While there is cold storage available for use by funds on the Lightning Network, its nodes are required to be online all the time to send and receive payments. The involved parties must remain online and use private keys to sign in. The computer storing the private keys is vulnerable to hacking.

If a user is offline for an extended period, they are also susceptible. As stated above, the leaving side is required to wait for 1,000 block confirmations to receive their deposit. If one side closes the channel while the other is away for eight days or more, this is a fraudulent channel close, and nothing can be done to recover the lost deposit. 

Network congestion due to a malicious attack also renders participants vulnerable to unreturned deposits. A forced expiration of many transactions could result and is a systemic risk recognized in the Lightning Network’s white paper.

A malicious party could overwhelm the capacity of a block by creating numerous channels that all expire at the same time. This situation would give the attacker the ability to steal funds from parties unable to withdraw them due to congestion. 

An Unstable Bitcoin Price

While the Lightning Network is supposed to make small transactions possible, there is still ample room for growth and improvement. The current increase in on-chain transactions is from trading volume. 

The added attention adds growth but volatility as well to the price. This volatility makes using Bitcoin as a payment tool difficult for merchants dealing with static inventory and planning accordingly with suppliers. When a company receives an invoice, it is often given 30 days or less to pay. If Bitcoin’s price drops 17.58% in 30 days, the supplier will be receiving an equivalent of 17.58% less fiat currency for their products. 

Graph courtesy of Google Finance

This could create the need for an entire market of futures for nearly every product, or for a “frozen” market price detailed on an invoice. A similar risk exists for consumers who use bitcoins to purchase goods or service while receiving incomes in other currencies.

The Lightning Network’s Future

As of January 2022, the Lightning Network had a capacity of 3,300 BTC, up from 2,000 in August 2021.  Arcane Research believes that up to 700 million users could be on the Lightning network by 2030.

The decentralized finance industry may help with this acceptance and use. Kraken stated at the end of 2020 that it would be making use of the Lightning Network in 2021:

We expect to allow clients to withdraw and deposit Bitcoin on Lightning in the first half of 2021, which will allow clients to move their Bitcoin instantly and with the lowest fees.

There may also be a solution for the automatic fraudulent channel close we discussed earlier in this article. “Watchtowers,” which are third parties running on Lightning Network nodes, could monitor off-chain transactions and prevent such transaction closes.


The Lightning Network represents a solid solution to a few of Bitcoin’s most significant issues. First and foremost, it encourages smaller transactions and widespread use.

If the security issues and malicious susceptibilities Lightning has can be solved, it will be a great addition to Bitcoin. More work is needed to improve Lightning. Yet it does give Bitcoin the opportunity of operating as a staple currency worldwide.  

Disclaimer:  The author of this text, Jean Chalopin, is a global business leader with a background encompassing banking, biotech, and entertainment. Mr. Chalopin is Chairman of Deltec International Group,

The co-author of this text, Robin Trehan, has a bachelor’s degree in economics, a master’s in international business and finance, and an MBA in electronic business. Mr. Trehan is a Senior VP at Deltec International Group,

The views, thoughts, and opinions expressed in this text are solely the views of the authors, and do not necessarily reflect those of Deltec International Group, its subsidiaries, and/or its employees.

AI and Aviation

The International Civil Aviation Organization (ICAO) stated that passenger numbers in 2021 were down by 49% from 2019, resulting in an expected loss of gross airline passenger operating revenues of $324 billion. But the market is expected to rebound in 2022, with an anticipated rise in passengers by up to 47%.

The previous two years demonstrated that neither airlines nor MRO (maintenance, repair, and overhaul) companies may depend on their previous triumphs. As a result, they face an unstated mandate to proactively future-proof their companies from top to bottom.

There are several ways AI, data, and technology can help the aviation industry recover.

Predictive Maintenance

The rapid adoption of predictive maintenance in the aviation industry remains a key driver of digital transformation for 2022.

Prior to the pandemic, only the top 10% of airlines implemented predictive maintenance. Yet in a recent opinion polls conducted by IFS, predictive maintenance revealed itself as the most frequently cited benefit of digital transformation.


Predictive maintenance forecasting algorithms–powered by AI (artificial intelligence)—are used by original equipment manufacturers (OEMs), airlines, or MROs to gather real-time and accurate data about every onboard system and sensor-connected component in their fleet.

For engines, which are relatively self-contained units, this practice is well established and produces a 30% performance efficiency gain over traditional techniques. It allows for a maintenance team to efficiently service and maintain aircraft components while maximizing their life spans.

As more flights arrive on time and without incident, this will benefit an airline’s passengers and its bottom line. There will likely be a significant increase in the number of airlines utilizing predictive maintenance soon.

Data and Analytics

The theme is to use data in new ways and enable organizations to better understand how well they are serving their customers. Data and analytics help determine the allocations of funds and budgets.

Given Covid-19, one of the most vital topics remains aeroplane air quality even as the virus abates in parts of the world. Managers consider new data as necessary. Data freshness will continue to be crucial for competing after the virus.

Platforms such as Google Cloud’s AI and ML technologies interpret data in timeframes that enable real-time decision-making. Data alone is useful, but sometimes insufficient. With sufficient context, data becomes powerful tools for strategic planning. 

For example, AirAsia uses Google Cloud to optimize pricing, improve revenue, and enhance the customer experience.

AirAsia started using an AI Platform in March 2018 to sort and predict demand for ancillary services such as baggage, seats, and meals, laying the groundwork for using machine learning to optimize pricing across a range of services. Other features include a digital health pass powered by AI.


Operational Performance

Passenger processing now bears a consistently larger impact on departure times due to the pandemic.

Travel restrictions, screening procedures, and available spaces frequently change, sometimes resulting in chaos. Passenger flow patterns can be modelled using machine learning to predict gate arrivals, passenger crowding, and the varying times it takes to leave different airports. Automated systems can reduce late gate arrivals and improve turnaround times.

Recently, the University of Cincinnati and Cincinnati/Northern Kentucky International Airport (CVG) announced that they are teaming up to predict (and reduce) crowding and improve the passenger experience.

Generative Design

Aerospace engineers and designers create components using generative design principles and AI. These principles give AI a set of parameters and enables it to generate a few possible designs. Then the results are manually improved upon for a final product.  


These more efficient components are then quickly created using artificial intelligence and machine learning techniques that learn from the guidelines laid out by the designers.

When it comes to creating new designs, generative design uses machine learning logic. Parametric modelling design in CAD software lags AI. Using generative design software yields a variety of possible solution combinations made possible by simply entering the relevant design parameters. The result is often thousands of variations of the same design, each achieving iteratively better outcomes. 

Fuel Efficiency

Even a tiny reduction in aircraft fuel consumptions significantly impacts a company’s bottom line and emissions through the power of volume. Aerospace companies place great importance upon fuel quality.

240 litres of fuel are used every second, and 14,400 pounds of fuel are used per hour on a typical commercial flight. We can reduce fuel consumption from 5% to 7% with the help of AI.

Practices powered by AI reduce fuel consumption. For example, machine learning aids pilots in optimizing their climb profiles before each flight. Safety Line data shows each flight can save 5% to 6% of its climb fuel without compromising passenger safety or comfort.

When applied to an airline fleet, this could reduce CO2 emissions by several thousand tonnes per year and operational costs by several million dollars. Optimizing the ascent process alone saves a staggering amount of fuel.

Customer Experience

Commercial aviation places a high value on customer satisfaction and service quality. Artificial intelligence in the airline industry improves customer service and engagement. Automated platforms powered by AI converse with customers in real-time and with human-like manners. Online chatbots save customers both time and effort by automating customer service processes such as:

  • Automating flight searches and bookings
  • Updating flights
  • Assisting with check-ins
  • Refunding flights and processing claims

Examples abound in the airline industry. Gol is the leading Brazilian airline and one of South America’s most important airlines. Gal, the company’s chatbot, assists passengers in booking and modifying flights, as well as with checking flight statuses.


Around 35% of Gol customers finalize their check-in using WhatsApp.

Keeping customers happy remains one of the airline industry’s biggest challenges. One bad experience causes a passenger to switch to a rival airline.

Customers who remain loyal to one airline may do so because of cost or convenience, or for miles they’ve accrued. All airlines prefer to see an increase in revenue per passenger and increased customer loyalty.

When it comes to increasing airline revenue per customer, airlines have traditionally relied on direct marketing and promotions. However, thanks to advances in AI, customer service and sales functions are mergeable.

In the event of a customer service call, the AI can recommend future trips or flights based upon the passenger’s past travel patterns (once the issue is resolved). After the service interaction, the AI may wait for days or weeks for the best time to contact the passenger, based on its analysis.


Artificial intelligence is beginning to take off in the aviation industry. Compared to other technologies, this one is still in its infancy. As its adoption spreads however, we’ll undoubtedly see more uses of it. Right now, we’re only seeing the beginning, and we’re eager to watch how it all plays out.

Disclaimer:  The author of this text, Jean Chalopin, is a global business leader with a background encompassing banking, biotech, and entertainment. Mr. Chalopin is Chairman of Deltec International Group,

The co-author of this text, Robin Trehan, has a bachelor’s degree in economics, a master’s in international business and finance, and an MBA in electronic business. Mr. Trehan is a Senior VP at Deltec International Group,

The views, thoughts, and opinions expressed in this text are solely the views of the authors, and do not necessarily reflect those of Deltec International Group, its subsidiaries, and/or its employees.

design and development by