I have no doubt: Machine learning (ML) is going to change how businesses make decisions, at almost every level. And this is especially true where I’ve spent most of my career — at the intersection of data analytics and supply chain management (SCM).
“We’ve arrived at a tipping point moment: The cost and effort of creating a machine learning SCM solution is far exceeded by the benefit of implementing it.”
I recently approached a friend at an iconic American brand to talk about my thoughts on how machine learning is going to revolutionize business. Now, this is someone I’ve worked with for years — a very smart guy. “How is machine learning going to be used?” he asked. “Is this going to sit in the C-suite and predict the next recession?”
“No,” I told him. “At least not for a long time.” Hyperbolic stories in mainstream media tell us that AI, in both good and evil form, is taking over the world. But Skynet — the truly thinking AI we all remember from Terminator — is a long way off.
However, I told him, “more focussed and massively impactful AI is here. Now. Waiting for you to use. And machine learning is one form of it.”
Since 2006, we’ve been developing tools that transform our clients’ raw data into insights and better decisions. That’s more than 12 years we’ve been successfully building business intelligence, creating custom dashboards and delivering key insights to analysts and supply chain professionals.
To do so, we’ve become experts at finding and applying the right tools to solve tough data-related business problems. Until recently, we’ve been wielding swords, with blades of varying lengths and shapes. Machine learning is our lightsaber. It adds another dimension to the business intelligence we deliver — and we’re betting big on it.
ML has the power to reduce the number of people needed to solve certain problems from hundreds to handfuls. Businesses that aren’t looking to adopt the technology at almost every decision point are headed the way of Sears and Gymboree.
Let me explain.
Why machine learning matters in supply chain management
“Machine learning is at the core of our journey towards artificial general intelligence, and in the meantime, it will change every industry and have a massive impact on our day-to-day lives.” —Vishal Maini & Samer Sabri, Machine Learning for Humans
In processing pure masses of data, machine learning has almost infinite capacity. And it can much more effectively utilize more types of data sources than traditional analytics techniques. Machine learning can also identify loose or intertwined data relationships, or patterns, that humans would never see. And in doing so, it takes on a predictive capacity. ML algorithms use real-time conditions to predict outcomes based on the relationship of conditions to outcomes. All the while it can be truly learning, so those predictions only get better.
“While the tools for applying machine learning are maturing fast, organizations still need professionals who are both versed in the technology and can understand business problems to successfully apply it.”
So what does this mean for supply chain management? (Editor’s note: We’ll be writing more posts about what machine learning looks like when applied skillfully. So, subscribe to our blog, and follow us on LinkedIn and Twitter if you’d like to join our conversation.)
In short, companies that effectively take advantage of machine learning will be able to make better, faster decisions while reducing staff and costs. The result will be huge top and bottom-line growth. For consumer companies, those reduced costs might even get passed down to consumer – fueling greater price competition for many goods. Companies that get left behind won’t be able to compete on price without decimating their bottom line.
Why machine learning now?
“[We are entering the wave of] business A.I. … Here, algorithms can be trained on proprietary data sets ranging from customer purchases to machine maintenance records to complex business processes—and ultimately lead managers to improved decision-making.” — Kai-Fu Lee, venture capitalist
Machine learning, of course, is hardly new. The term dates back to 1959, when IBM’s Arthur Lee Samuel coined the phrase. In recent years, the world’s largest tech companies — e.g. Google, Apple, Amazon, and Microsoft — have been investing billions in AI and machine learning. Meanwhile, an ecosystem of startups rooted in data science have been driving innovation.
However, practical applications of machine learning have been sparse. For emerging technologies on the cusp of mass adoption, it often takes some kind of enabler or event to take them the next level.
For machine learning and AI, that enabler is the emergence of platforms and open-source libraries. These production-ready tools can be customized and applied to a business problem. Today, the machine learning tools available for savvy developers — coming from cloud services like AWS and Azure to startups like H2O.ai — is maturing fast.
With the availability of these tools, we’ve arrived at a tipping point moment: The cost and effort of creating a machine learning SCM solution is far exceeded by the benefit of implementing it.
BUT, that doesn’t mean that the application of machine learning is easy.
Enter eAlchemy — solving complex business problems by finding the right tools
“Machine learning lets us solve classes of problem that computers could not usefully address before, but each of those problems will require a different implementation, and different data, a different route to market, and often a different company.” — Benedict Evans, Andreessen Horowitz
While the tools for applying machine learning are maturing fast, organizations still need professionals who are both versed in the technology and can understand business problems to successfully apply it.
Consider the 3D printer. The cost of the technology behind the machines has decreased dramatically over the past decade. And it’s spurred a wave of hobbyist hardware makers around the world. But you can’t just talk to a 3D printer and tell it to build, for example, a new cabinet handle. You have to understand how the handle fits in the cabinet door, determine the size and color you want, and then precisely design the three-dimensional piece using software for the printer.
Machine learning isn’t push-button magic. You can’t just drop a code string into an ERP and instantly get smarter about your supply chain. You need to understand the business problem you are trying to solve and the outcome you are trying to achieve. And you need the development expertise to choose, configure, apply, and train the machine learning algorithm.
That’s why, as we head deeper into 2019, I’m putting a lot more resources into our machine learning practice area. As I mentioned, we’ve been providing insight to operations and supply chain teams since 2006. But we’re more than just software developers or data scientists. We are experts at finding the right tools and applying them to create business value. And machine learning is the most powerful tool we’ve ever wielded.
Not a sword, a lightsaber.
What can machine learning do for you?
If you’d like to learn more about how your business can benefit from machine learning, we’re happy to offer up advice. Contact us to set up a time to talk.