Machine Learning And Business Problem-Solving

David Cruickshank

2018 at the SAP Co-Innovation Lab in Silicon Valley is off to a roaring start, with three of its projects heavily focused on the topic of artificial intelligence. This project work spans machine learning, cognitive computing, computer vision, and even touches the realms of natural-language processing.

For our lab, we began digging into the application of machine learning beginning in 2014, exploring its application in everything from supply chain optimization to factory automation and retail, including predicting terrorist attacks. Where we can apply knowledge for a given domain and weave it into a learning algorithm for the sake of doing non-deterministic pattern recognition, machine learning grounded in only statistics (not symbology, logic, or evolutionary) can readily improve upon guessing. Learning from a productive data set, and where overfitting is sufficiently avoided or mitigated, a learning algorithm can recognize patterns and generalize to cases not yet encountered. Such explorations for us started more than two years ago with SAP NS2 and ConvergentAI (formerly AxxonAI) where we find the project team’s proof-of-concept (POC) results remain relevant today, but applicable to problem-solving the same way in other domains.

While conceptually different, a strong relationship exists between machine learning and analytics where machine learning uses data and learning algorithms (supervised and unsupervised) to optimize a model based on performance and prior experience. The resulting model is often used to improve the accuracy of analytics. Predictive analytics uses this learned model to find patterns against new data used to make informed predictions about future events.

Applying swarm intelligence for a POC featuring event risk forecasting

In the co-innovation lab, the project team worked to integrate what ConvergentAI developed as a unique predictive analytics capability, combining a decentralized and continuous machine learning engine with a decentralized and continuous analytic forecasting model. ConvergentAI calls its predictive analytics “swarm intelligence.”

SAP NS2 is a part of the family of SAP companies. One area of focus is working with U.S. national security agencies. In its collaborative project work both within SAP and with select partners from its ecosystem, SAP NS2 brought a data fusion platform together in a way that cleverly incorporates the machine learning and analytics capabilities of ConvergentAI to provide an event risk forecasting capability.

This application of event risk forecasting is designed to reduce uncertainty or risk with a process model that connects past events to their possible origins to analyze and predict future events. Its general reasoning and learning approach can be applied to a wide range of domains.

Key to this integrated capability is the ability to ingest, exploit, make sense of, and create a knowledge base of information that can be used as the source for event risk forecasting. Event risk forecasting requires conditioned information in order to drive the swarming model to forecast future events.

SAP NS2 developed its data fusion architecture on SAP HANA as a proof of concept to support a general-purpose, closed-loop process to integrate and correlate data, and thus create a knowledge base within an in-memory database for exploration and discovery of information. The database also serves as the source for other applications such as the event risk forecasting engine developed by ConvergentAI.

Continuously updated risk forecasting

The swarm intelligence technology and corresponding models become meaningful to a human analyst in explaining why a particular output is generated. The event risk forecasting engine runs continuously, enabling such changes to be translated into changed risk forecasts, building the human operator’s intuition of the correspondence of domain assumptions to risk distributions. In contrast to many other machine learning applications, the event risk forecasting engine continuously reevaluates its assertions and conclusions based on currently available data and parameters.

Broad industry application

The co-innovation lab project fully demonstrated intelligent data fusion and event risk forecasting in an effort to predict future terror activity or a given actor’s relationship to an act of terror using a substantial data set of event violence in Africa. From this demo, it is easy to see how this capability can be applied domestically in law enforcement (smarter allocation of police), and also commercially for the protection of large-scale critical infrastructure (oil and gas operations, airfields, or other transportation). Beyond forecasting the probability of events with the intent to suppress them, industries such as insurance have an interest in predicting the risk for contingency planning purposes.

For a more complete understanding of our work on event risk forecasting at SAP Co-Innovation Lab, we invite you take a look at our white paper “Predictive Analytics with Intelligent Data Fusion Event Risk Forecasting.” To learn more or to possibly explore how this can be applied to your business, please comment and/or connect with us to explore this in richer detail.


David Cruickshank

About David Cruickshank

David Cruickshank is senior director for strategy and operations for the SAP Co-Innovation Lab. He leads the lab's efforts in Silicon Valley to enable ecosystem-driven co-innovation between SAP, its partners, and customers. Additionally, he manages all operational aspects necessary to run a multimillion-dollar data center to provision private cloud infrastructures to deliver productive SAP landscapes consumed by co-innovation projects seeking a faster track to market for commercially successful innovations.

Using Machine Learning To Turbocharge Financial Services Innovation

Toni Tomic

When people talk about artificial intelligence (AI), the discussion commonly turns to flashy robotics – how they support manufacturing production lines, disable explosives, or even vacuum the floors.

Major steps in AI were made in the late 1950s and early ‘60s, with flagship examples like the ELIZA computer program that demonstrated the superficiality of communication between humans and machines. From then until the 1980s, there was great promise that AI could revolutionize businesses, but there was no major disruption.

Today it feels as if AI is born again, and it is much more than robots. Innovative new AI technologies are delivering benefits to a wide variety of industries, including financial services.

According to the “Worldwide Semiannual Cognitive Artificial Intelligence Systems Spending Guide” from International Data Corporation (IDC), worldwide revenues for cognitive and artificial intelligence (AI) systems will reach $12.0 billion in 2017, an increase of 59.1% over 2016. “Cognitive and artificial intelligence solutions continue to proliferate across all industries, resulting in significant growth opportunities,” said Marianne Daquila, research manager, Customer Insights and Analysis, at IDC. “Some of the use cases are very industry specific, such as diagnosis and treatment in healthcare, and in others, they are common across multiple industries such as automated customer service agents. The variety, application, and nature of cognitive/artificial intelligence use cases are resulting in ubiquitous spend over the forecast period.”

One of the most interesting disciplines is machine learning, a specific type of AI that allows computers to learn without being explicitly programmed to do so. Machine learning uses statistical theory and exponentially more powerful computer processing to help businesses quickly realize valuable insight from their data.

This is great news for banks and insurers. Not only are these service providers facing falling profit margins, rising customer expectations, and increasing competition from fintechs, but they also need to cut costs. With machine learning, they can extract value from huge volumes of data, cheaply and effectively.

Machine learning can help traditional global banks that operate accounts at 140 to 170 British pound sterling (GBP) to compete better with challenger banks that operate at 4 to 44 GBP.

Creating intelligent financial services

Machine learning is ideal for addressing three dominant financial services challenges (see figure), including:

  • Customer front office: Unsupervised machine learning techniques can help banks and insurers segment their customers and offer personalized, targeted products. These technologies can also improve speed and agility, helping companies compete with fintech firms through enhanced knowledge of their customers.
  • Regulation and compliance: Using automated reports, stress-testing solutions, and behavioral analysis of e-mails and phone records to identify suspicious customer or employee behavior, machine learning can boost regulatory compliance. It can also enhance fraud detection, improve anti-money laundering efforts, and more effectively detect credit risk.
  • Operational efficiency: By combining Big Data with machine learning, financial services companies can automate back-office operations, reduce errors, and accelerate process execution. Insurers can improve and automate claims handling by recognizing patterns in pictures or individuals involved in damages, for example. Machine learning algorithms can also elevate talent management and recruitment by evaluating the resumes of successful employees while searching for online candidates with similar traits and experience.

Financial services challenges addressed by machine learning

1. Automating front-office applications

One of the most interesting machine learning applications helps financial services companies segment customers and offer targeted products or services. Let’s look at how the technology works.

Cluster analysis discovers distinct groups within the customer base and identifies similarities over several dimensions. Because this process is unsupervised, banks or insurers do not need to define the characteristics. The technology discovers these on its own.

Once the customer base is segmented, the technology builds predictive models. Algorithms help identify the most suitable products for each customer. And because the algorithms learn as they go, they can recognize changes in customer preferences in real time and automatically adjust product recommendations and provide the right advice at the right time.

The benefits can be significant. Personalized offerings make customers feel understood, increasing satisfaction. Successful cross-sell and upsell efforts can increase revenues. Service speed increases when banks and insurers can automatically recognize a change in behavior and respond instantly, without human intervention.

2. Choosing the right AI functionality

Traditional database technologies and analytics tools are not powerful enough to support machine learning. Fortunately, a new generation of innovative solutions is coming to market.

Many vendors are seeking to capitalize on this burgeoning industry. As with any new technology, decision-makers must carefully assess how well the tools meet the needs of the business.

When choosing a machine learning solution for customer retention, for example, we advise financial services companies to select tools that automatically:

  • Manage dynamic data from a variety of customer channels and build an overview of the customer journey
  • Sort, classify, and route events, pinpointing critical changes and reliable customer churn indicators
  • Identify customers who are about to churn and take proactive steps to prevent customers from leaving

In the battle to provide better customer service, machine learning technology is becoming a differentiating technology for financial services institutions. There are two ways to get there. Companies can:

  • Use standard applications, where machine learning is embedded and shipped with the software
  • Develop their own applications based on available cloud platform solutions and toolsets

Preparing to address obstacles

Without a doubt, there is huge potential to bring efficiency and effectiveness to a new level by injecting AI and machine learning into the financial services business. So what is hindering financial services institutions from becoming more successful with AI and machine learning and gaining benefits from disruptive technology?

There are three major obstacles:

  • Full senior management buy-in that extends beyond funding a proof of concept
  • Shortage of specialist skills to operate and maintain the technology
  • Costs of the AI system

With these challenges, getting quick wins in a short time frame is becoming more and more crucial. And since vendors are embedding machine learning scenarios into existing software applications, this might be a good starting point for companies that want to simultaneously work on major disruptive use cases.

Is your company ready to embrace the competitive advantages offered by AI and machine learning? To learn more about how machine learning can help financial services companies innovate and compete, contact me in the comments below or on LinkedIn.


Toni Tomic

About Toni Tomic

Toni Tomic is the Vice President and Global Head of Transformation at SAP. He is responsible for the SAP innovation strategy for disruptive technology such as artificial intelligence, blockchain, and the Internet of Things. Toni also oversees FinTechs and drives financial services business development globally. Before joining SAP, Toni worked for 10 years in strategy and management consulting focused on business and IT strategy, post-merger integration, and restructuring programs in the financial services industry.

Air Cover For The Endangered

Rick Price

David Allen sits in a semi-dark trailer, eyes on a computer screen. He is scanning an image from an aerial drone skimming the treetops at the Dinokeng Game Reserve in South Africa. It is searching for elephants. A cloud of dust appears in the camera’s field of view – the plane banks toward it, and there! A big bull, ears flapping with each step, a tracking collar around his neck. The ERP Air Force has found its target, with the aim not to destroy, but to save the creature from his nemesis, the human poacher.

The link to poverty

The group behind this – a non-governmental organization called Elephants, Rhinos, & People or ERP, was founded in 2014 by to preserve and protect Southern Africa’s wild elephants and rhinos through alleviating rural poverty. A look at some numbers shows the link between poverty and threats to the elephants and rhinos.

According to, rhino horn is worth up to hundreds of times the per-capita income in poor rural areas of Southern Africa. On the global black market, it can fetch up to $80,000 per kilogram. Depending on the species, a rhino horn can weigh three to four kilos, so one horn can command more than $300,000 – an unimaginable fortune to most people there.

Grim statistics

The result is carnage. Wild African elephants are being killed at the rate of four an hour. At the end of 2016, the population was 352,000, according to Group Elephant. Rhinos are in far worse shape: The total population as of this writing is 19,682 Southern White Rhinos and 5,042 Black Rhinos for a total of 24,724 in all of Africa. The government of South Africa says 529 rhinos were poached in that nation alone in the first half of 2017. ERP says all in all, three rhinos are killed every day. Any effort to preserve this species must give people an alternative to poaching and a stake in the animals’ survival, through jobs and economic opportunities like tourism. Conservation initiatives also have to make the most of scarce resources, and that requires cutting-edge technology.

Covering a lot of ground

These are big animals, they can move quickly, and their stomping grounds are large. Dinokeng alone encompasses more than 45,000 acres. It would be impractical to hire enough people to keep tabs on every elephant and every rhino there every moment of every day. Inside the reserve, they are safer, but if they break out into the surrounding suburban area, just north of Pretoria, they face threats, including exposure to poachers who might not risk the reserve itself. ERP works with Dinokeng to patrol the fences and help keep the animals in, but there’s a limit to what they can accomplish. That’s where the ERP Air Force, the tracking collars, and Big Data come in.

Eyes in the sky

Back in the trailer, the GPS collar monitoring system has alerted drone pilot David Allen that the bull is getting too close to the fences. The drone follows the elephants as they approach the limits of the reserve. The operator can then direct rangers in SUVs or a helicopter to nudge the herd away from the boundary. That reduces the poachers’ ability to kill the animals – outside Dinokeng, there is less protection. The drones will also watch for poaching within the reserve, and ERP hopes to use them for other initiatives.

Big Data to protect big creatures

ERP uses cutting-edge digital technology to save these wild elephants and rhinos, both by keeping them in safe areas and by alleviating the poverty of the local population. All of that requires the ability to wrangle huge amounts of data quickly, from knowing where the elephants are, to routing the people to them. A group of technology companies, including SAP and its partner EPI-USE, have collaborated with ERP to build a system capable of collecting, organizing, storing, and retrieving that information.

Next on the roadmap is predictive analytics: As EPI-USE’s Jan van Rensburg said in May 2017, “We rely upon analytics in the background, a platform where we can predict where poaching will happen. And we use the data that we build up and store in our in-memory database so that we can more efficiently use the drones, and send them to the spots where poachers are more likely to be. And when a poacher is found by the drone, we can dispatch a ranger to deal with it.”

A weapon to save, not kill

So for ERP, Big Data and the ability to use it in real time have become a weapon in the fight to save endangered elephants and rhinos, while helping people in poverty create new ways to earn a living and turn away from poaching.

Want to know more about how ERP uses Big Data and drones to save endangered wild Elephants and Rhinos through alleviating rural poverty? Watch the video.


Rick Price

About Rick Price

Rick Price is an Emmy Award-winning journalist who now works at SAP, where he tells stories of customers' digital transformation.

The Blockchain Solution

By Gil Perez, Tom Raftery, Hans Thalbauer, Dan Wellers, and Fawn Fitter

In 2013, several UK supermarket chains discovered that products they were selling as beef were actually made at least partly—and in some cases, entirely—from horsemeat. The resulting uproar led to a series of product recalls, prompted stricter food testing, and spurred the European food industry to take a closer look at how unlabeled or mislabeled ingredients were finding their way into the food chain.

By 2020, a scandal like this will be eminently preventable.

The separation between bovine and equine will become immutable with Internet of Things (IoT) sensors, which will track the provenance and identity of every animal from stall to store, adding the data to a blockchain that anyone can check but no one can alter.

Food processing companies will be able to use that blockchain to confirm and label the contents of their products accordingly—down to the specific farms and animals represented in every individual package. That level of detail may be too much information for shoppers, but they will at least be able to trust that their meatballs come from the appropriate species.

The Spine of Digitalization

Keeping food safer and more traceable is just the beginning, however. Improvements in the supply chain, which have been incremental for decades despite billions of dollars of technology investments, are about to go exponential. Emerging technologies are converging to transform the supply chain from tactical to strategic, from an easily replicable commodity to a new source of competitive differentiation.

You may already be thinking about how to take advantage of blockchain technology, which makes data and transactions immutable, transparent, and verifiable (see “What Is Blockchain and How Does It Work?”). That will be a powerful tool to boost supply chain speed and efficiency—always a worthy goal, but hardly a disruptive one.

However, if you think of blockchain as the spine of digitalization and technologies such as AI, the IoT, 3D printing, autonomous vehicles, and drones as the limbs, you have a powerful supply chain body that can leapfrog ahead of its competition.

What Is Blockchain and How Does It Work?

Here’s why blockchain technology is critical to transforming the supply chain.

Blockchain is essentially a sequential, distributed ledger of transactions that is constantly updated on a global network of computers. The ownership and history of a transaction is embedded in the blockchain at the transaction’s earliest stages and verified at every subsequent stage.

A blockchain network uses vast amounts of computing power to encrypt the ledger as it’s being written. This makes it possible for every computer in the network to verify the transactions safely and transparently. The more organizations that participate in the ledger, the more complex and secure the encryption becomes, making it increasingly tamperproof.

Why does blockchain matter for the supply chain?

  • It enables the safe exchange of value without a central verifying partner, which makes transactions faster and less expensive.
  • It dramatically simplifies recordkeeping by establishing a single, authoritative view of the truth across all parties.
  • It builds a secure, immutable history and chain of custody as different parties handle the items being shipped, and it updates the relevant documentation.
  • By doing these things, blockchain allows companies to create smart contracts based on programmable business logic, which can execute themselves autonomously and thereby save time and money by reducing friction and intermediaries.

Hints of the Future

In the mid-1990s, when the World Wide Web was in its infancy, we had no idea that the internet would become so large and pervasive, nor that we’d find a way to carry it all in our pockets on small slabs of glass.

But we could tell that it had vast potential.

Today, with the combination of emerging technologies that promise to turbocharge digital transformation, we’re just beginning to see how we might turn the supply chain into a source of competitive advantage (see “What’s the Magic Combination?”).

What’s the Magic Combination?

Those who focus on blockchain in isolation will miss out on a much bigger supply chain opportunity.

Many experts believe emerging technologies will work with blockchain to digitalize the supply chain and create new business models:

  • Blockchain will provide the foundation of automated trust for all parties in the supply chain.
  • The IoT will link objects—from tiny devices to large machines—and generate data about status, locations, and transactions that will be recorded on the blockchain.
  • 3D printing will extend the supply chain to the customer’s doorstep with hyperlocal manufacturing of parts and products with IoT sensors built into the items and/or their packaging. Every manufactured object will be smart, connected, and able to communicate so that it can be tracked and traced as needed.
  • Big Data management tools will process all the information streaming in around the clock from IoT sensors.
  • AI and machine learning will analyze this enormous amount of data to reveal patterns and enable true predictability in every area of the supply chain.

Combining these technologies with powerful analytics tools to predict trends will make lack of visibility into the supply chain a thing of the past. Organizations will be able to examine a single machine across its entire lifecycle and identify areas where they can improve performance and increase return on investment. They’ll be able to follow and monitor every component of a product, from design through delivery and service. They’ll be able to trigger and track automated actions between and among partners and customers to provide customized transactions in real time based on real data.

After decades of talk about markets of one, companies will finally have the power to create them—at scale and profitably.

Amazon, for example, is becoming as much a logistics company as a retailer. Its ordering and delivery systems are so streamlined that its customers can launch and complete a same-day transaction with a push of a single IP-enabled button or a word to its ever-attentive AI device, Alexa. And this level of experimentation and innovation is bubbling up across industries.

Consider manufacturing, where the IoT is transforming automation inside already highly automated factories. Machine-to-machine communication is enabling robots to set up, provision, and unload equipment quickly and accurately with minimal human intervention. Meanwhile, sensors across the factory floor are already capable of gathering such information as how often each machine needs maintenance or how much raw material to order given current production trends.

Once they harvest enough data, businesses will be able to feed it through machine learning algorithms to identify trends that forecast future outcomes. At that point, the supply chain will start to become both automated and predictive. We’ll begin to see business models that include proactively scheduling maintenance, replacing parts just before they’re likely to break, and automatically ordering materials and initiating customer shipments.

Italian train operator Trenitalia, for example, has put IoT sensors on its locomotives and passenger cars and is using analytics and in-memory computing to gauge the health of its trains in real time, according to an article in Computer Weekly. “It is now possible to affordably collect huge amounts of data from hundreds of sensors in a single train, analyse that data in real time and detect problems before they actually happen,” Trenitalia’s CIO Danilo Gismondi told Computer Weekly.

Blockchain allows all the critical steps of the supply chain to go electronic and become irrefutably verifiable by all the critical parties within minutes: the seller and buyer, banks, logistics carriers, and import and export officials.

The project, which is scheduled to be completed in 2018, will change Trenitalia’s business model, allowing it to schedule more trips and make each one more profitable. The railway company will be able to better plan parts inventories and determine which lines are consistently performing poorly and need upgrades. The new system will save €100 million a year, according to ARC Advisory Group.

New business models continue to evolve as 3D printers become more sophisticated and affordable, making it possible to move the end of the supply chain closer to the customer. Companies can design parts and products in materials ranging from carbon fiber to chocolate and then print those items in their warehouse, at a conveniently located third-party vendor, or even on the client’s premises.

In addition to minimizing their shipping expenses and reducing fulfillment time, companies will be able to offer more personalized or customized items affordably in small quantities. For example, clothing retailer Ministry of Supply recently installed a 3D printer at its Boston store that enables it to make an article of clothing to a customer’s specifications in under 90 minutes, according to an article in Forbes.

This kind of highly distributed manufacturing has potential across many industries. It could even create a market for secure manufacturing for highly regulated sectors, allowing a manufacturer to transmit encrypted templates to printers in tightly protected locations, for example.

Meanwhile, organizations are investigating ways of using blockchain technology to authenticate, track and trace, automate, and otherwise manage transactions and interactions, both internally and within their vendor and customer networks. The ability to collect data, record it on the blockchain for immediate verification, and make that trustworthy data available for any application delivers indisputable value in any business context. The supply chain will be no exception.

Blockchain Is the Change Driver

The supply chain is configured as we know it today because it’s impossible to create a contract that accounts for every possible contingency. Consider cross-border financial transfers, which are so complex and must meet so many regulations that they require a tremendous number of intermediaries to plug the gaps: lawyers, accountants, customer service reps, warehouse operators, bankers, and more. By reducing that complexity, blockchain technology makes intermediaries less necessary—a transformation that is revolutionary even when measured only in cost savings.

“If you’re selling 100 items a minute, 24 hours a day, reducing the cost of the supply chain by just $1 per item saves you more than $52.5 million a year,” notes Dirk Lonser, SAP go-to-market leader at DXC Technology, an IT services company. “By replacing manual processes and multiple peer-to-peer connections through fax or e-mail with a single medium where everyone can exchange verified information instantaneously, blockchain will boost profit margins exponentially without raising prices or even increasing individual productivity.”

But the potential for blockchain extends far beyond cost cutting and streamlining, says Irfan Khan, CEO of supply chain management consulting and systems integration firm Bristlecone, a Mahindra Group company. It will give companies ways to differentiate.

“Blockchain will let enterprises more accurately trace faulty parts or products from end users back to factories for recalls,” Khan says. “It will streamline supplier onboarding, contracting, and management by creating an integrated platform that the company’s entire network can access in real time. It will give vendors secure, transparent visibility into inventory 24×7. And at a time when counterfeiting is a real concern in multiple industries, it will make it easy for both retailers and customers to check product authenticity.”

Blockchain allows all the critical steps of the supply chain to go electronic and become irrefutably verifiable by all the critical parties within minutes: the seller and buyer, banks, logistics carriers, and import and export officials. Although the key parts of the process remain the same as in today’s analog supply chain, performing them electronically with blockchain technology shortens each stage from hours or days to seconds while eliminating reams of wasteful paperwork. With goods moving that quickly, companies have ample room for designing new business models around manufacturing, service, and delivery.

Challenges on the Path to Adoption

For all this to work, however, the data on the blockchain must be correct from the beginning. The pills, produce, or parts on the delivery truck need to be the same as the items listed on the manifest at the loading dock. Every use case assumes that the data is accurate—and that will only happen when everything that’s manufactured is smart, connected, and able to self-verify automatically with the help of machine learning tuned to detect errors and potential fraud.

Companies are already seeing the possibilities of applying this bundle of emerging technologies to the supply chain. IDC projects that by 2021, at least 25% of Forbes Global 2000 (G2000) companies will use blockchain services as a foundation for digital trust at scale; 30% of top global manufacturers and retailers will do so by 2020. IDC also predicts that by 2020, up to 10% of pilot and production blockchain-distributed ledgers will incorporate data from IoT sensors.

Despite IDC’s optimism, though, the biggest barrier to adoption is the early stage level of enterprise use cases, particularly around blockchain. Currently, the sole significant enterprise blockchain production system is the virtual currency Bitcoin, which has unfortunately been tainted by its associations with speculation, dubious financial transactions, and the so-called dark web.

The technology is still in a sufficiently early stage that there’s significant uncertainty about its ability to handle the massive amounts of data a global enterprise supply chain generates daily. Never mind that it’s completely unregulated, with no global standard. There’s also a critical global shortage of experts who can explain emerging technologies like blockchain, the IoT, and machine learning to nontechnology industries and educate organizations in how the technologies can improve their supply chain processes. Finally, there is concern about how blockchain’s complex algorithms gobble computing power—and electricity (see “Blockchain Blackouts”).

Blockchain Blackouts

Blockchain is a power glutton. Can technology mediate the issue?

A major concern today is the enormous carbon footprint of the networks creating and solving the algorithmic problems that keep blockchains secure. Although virtual currency enthusiasts claim the problem is overstated, Michael Reed, head of blockchain technology for Intel, has been widely quoted as saying that the energy demands of blockchains are a significant drain on the world’s electricity resources.

Indeed, Wired magazine has estimated that by July 2019, the Bitcoin network alone will require more energy than the entire United States currently uses and that by February 2020 it will use as much electricity as the entire world does today.

Still, computing power is becoming more energy efficient by the day and sticking with paperwork will become too slow, so experts—Intel’s Reed among them—consider this a solvable problem.

“We don’t know yet what the market will adopt. In a decade, it might be status quo or best practice, or it could be the next Betamax, a great technology for which there was no demand,” Lonser says. “Even highly regulated industries that need greater transparency in the entire supply chain are moving fairly slowly.”

Blockchain will require acceptance by a critical mass of companies, governments, and other organizations before it displaces paper documentation. It’s a chicken-and-egg issue: multiple companies need to adopt these technologies at the same time so they can build a blockchain to exchange information, yet getting multiple companies to do anything simultaneously is a challenge. Some early initiatives are already underway, though:

  • A London-based startup called Everledger is using blockchain and IoT technology to track the provenance, ownership, and lifecycles of valuable assets. The company began by tracking diamonds from mine to jewelry using roughly 200 different characteristics, with a goal of stopping both the demand for and the supply of “conflict diamonds”—diamonds mined in war zones and sold to finance insurgencies. It has since expanded to cover wine, artwork, and other high-value items to prevent fraud and verify authenticity.
  • In September 2017, SAP announced the creation of its SAP Leonardo Blockchain Co-Innovation program, a group of 27 enterprise customers interested in co-innovating around blockchain and creating business buy-in. The diverse group of participants includes management and technology services companies Capgemini and Deloitte, cosmetics company Natura Cosméticos S.A., and Moog Inc., a manufacturer of precision motion control systems.
  • Two of Europe’s largest shipping ports—Rotterdam and Antwerp—are working on blockchain projects to streamline interaction with port customers. The Antwerp terminal authority says eliminating paperwork could cut the costs of container transport by as much as 50%.
  • The Chinese online shopping behemoth Alibaba is experimenting with blockchain to verify the authenticity of food products and catch counterfeits before they endanger people’s health and lives.
  • Technology and transportation executives have teamed up to create the Blockchain in Transport Alliance (BiTA), a forum for developing blockchain standards and education for the freight industry.

It’s likely that the first blockchain-based enterprise supply chain use case will emerge in the next year among companies that see it as an opportunity to bolster their legal compliance and improve business processes. Once that happens, expect others to follow.

Customers Will Expect Change

It’s only a matter of time before the supply chain becomes a competitive driver. The question for today’s enterprises is how to prepare for the shift. Customers are going to expect constant, granular visibility into their transactions and faster, more customized service every step of the way. Organizations will need to be ready to meet those expectations.

If organizations have manual business processes that could never be automated before, now is the time to see if it’s possible. Organizations that have made initial investments in emerging technologies are looking at how their pilot projects are paying off and where they might extend to the supply chain. They are starting to think creatively about how to combine technologies to offer a product, service, or business model not possible before.

A manufacturer will load a self-driving truck with a 3D printer capable of creating a customer’s ordered item en route to delivering it. A vendor will capture the market for a socially responsible product by allowing its customers to track the product’s production and verify that none of its subcontractors use slave labor. And a supermarket chain will win over customers by persuading them that their choice of supermarket is also a choice between being certain of what’s in their food and simply hoping that what’s on the label matches what’s inside.

At that point, a smart supply chain won’t just be a competitive edge. It will become a competitive necessity. D!

About the Authors

Gil Perez is Senior Vice President, Internet of Things and Digital Supply Chain, at SAP.

Tom Raftery is Global Vice President, Futurist, and Internet of Things Evangelist, at SAP.

Hans Thalbauer is Senior Vice President, Internet of Things and Digital Supply Chain, at SAP.

Dan Wellers is Global Lead, Digital Futures, at SAP.

Fawn Fitter is a freelance writer specializing in business and technology.

Read more thought provoking articles in the latest issue of the Digitalist Magazine, Executive Quarterly.



The Differences Between Machine Learning And Predictive Analytics

Shaily Kumar

Many people are confused about the specifics of machine learning and predictive analytics. Although they are both centered on efficient data processing, there are many differences.

Machine learning

Machine learning is a method of computational learning underlying most artificial intelligence (AI) applications. In ML, systems or algorithms improve themselves through data experience without relying on explicit programming. ML algorithms are wide-ranging tools capable of carrying out predictions while simultaneously learning from over trillions of observations.

Machine learning is considered a modern-day extension of predictive analytics. Efficient pattern recognition and self-learning are the backbones of ML models, which automatically evolve based on changing patterns in order to enable appropriate actions.

Many companies today depend on machine learning algorithms to better understand their clients and potential revenue opportunities. Hundreds of existing and newly developed machine learning algorithms are applied to derive high-end predictions that guide real-time decisions with less reliance on human intervention.

Business application of machine learning: employee satisfaction

One common, uncomplicated, yet successful business application of machine learning is measuring real-time employee satisfaction.

Machine learning applications can be highly complex, but one that’s both simple and very useful for business is a machine learning algorithm that compares employee satisfaction ratings to salaries. Instead of plotting a predictive satisfaction curve against salary figures for various employees, as predictive analytics would suggest, the algorithm assimilates huge amounts of random training data upon entry, and the prediction results are affected by any added training data to produce real-time accuracy and more helpful predictions.

This machine learning algorithm employs self-learning and automated recalibration in response to pattern changes in the training data, making machine learning more reliable for real-time predictions than other AI concepts. Repeatedly increasing or updating the bulk of training data guarantees better predictions.

Machine learning can also be implemented in image classification and facial recognition with deep learning and neural network techniques.

Predictive analytics

Predictive analytics can be defined as the procedure of condensing huge volumes of data into information that humans can understand and use. Basic descriptive analytic techniques include averages and counts. Descriptive analytics based on obtaining information from past events has evolved into predictive analytics, which attempts to predict the future based on historical data.

This concept applies complex techniques of classical statistics, like regression and decision trees, to provide credible answers to queries such as: ‘’How exactly will my sales be influenced by a 10% increase in advertising expenditure?’’ This leads to simulations and “what-if” analyses for users to learn more.

All predictive analytics applications involve three fundamental components:

  • Data: The effectiveness of every predictive model strongly depends on the quality of the historical data it processes.
  • Statistical modeling: Includes the various statistical techniques ranging from basic to complex functions used for the derivation of meaning, insight, and inference. Regression is the most commonly used statistical technique.
  • Assumptions: The conclusions drawn from collected and analyzed data usually assume the future will follow a pattern related to the past.

Data analysis is crucial for any business en route to success, and predictive analytics can be applied in numerous ways to enhance business productivity. These include things like marketing campaign optimization, risk assessment, market analysis, and fraud detection.

Business application of predictive analytics: marketing campaign optimization

In the past, valuable marketing campaign resources were wasted by businesses using instincts alone to try to capture market niches. Today, many predictive analytic strategies help businesses identify, engage, and secure suitable markets for their services and products, driving greater efficiency into marketing campaigns.

A clear application is using visitors’ search history and usage patterns on e-commerce websites to make product recommendations. Sites like Amazon increase their chance of sales by recommending products based on specific consumer interests. Predictive analytics now plays a vital role in the marketing operations of real estate, insurance, retail, and almost every other sector.

How machine learning and predictive analytics are related

While businesses must understand the differences between machine learning and predictive analytics, it’s just as important to know how they are related. Basically, machine learning is a predictive analytics branch. Despite having similar aims and processes, there are two main differences between them:

  • Machine learning works out predictions and recalibrates models in real-time automatically after design. Meanwhile, predictive analytics works strictly on “cause” data and must be refreshed with “change” data.
  • Unlike machine learning, predictive analytics still relies on human experts to work out and test the associations between cause and outcome.

Explore machine learning applications and AI software with SAP Leonardo.


Shaily Kumar

About Shaily Kumar

Shailendra has been on a quest to help organisations make money out of data and has generated an incremental value of over one billion dollars through analytics and cognitive processes. With a global experience of more than two decades, Shailendra has worked with a myriad of Corporations, Consulting Services and Software Companies in various industries like Retail, Telecommunications, Financial Services and Travel - to help them realise incremental value hidden in zettabytes of data. He has published multiple articles in international journals about Analytics and Cognitive Solutions; and recently published “Making Money out of Data” which showcases five business stories from various industries on how successful companies make millions of dollars in incremental value using analytics. Prior to joining SAP, Shailendra was Partner / Analytics & Cognitive Leader, Asia at IBM where he drove the cognitive business across Asia. Before joining IBM, he was the Managing Director and Analytics Lead at Accenture delivering value to its clients across Australia and New Zealand. Coming from the industry, Shailendra held key Executive positions driving analytics at Woolworths and Coles in the past.