Skip to main content

Even before generative artificial intelligence (GenAI) exploded onto the scene late last year, artificial intelligence (AI) had been growing in importance for various supply chain management (SCM) applications. In fact, a recent survey indicates it’s making a big difference for top supply chain organizations. 

“Top performing supply chain organizations are investing in artificial intelligence and machine learning (AI/ML) to optimize their processes at more than twice the rate of low performing peers, according to a survey by Gartner, Inc.,” the press release says. “The survey also revealed that the best supply chain organizations are using productivity, rather than efficiency or cost savings, as their key focus to sustain business momentum over the next three years.”

Gartner’s survey of 818 supply chain practitioners “across geography and industry” in late 2023 aimed to gain deeper insights about “how the supply chain is adapting to changes in economic values, fostering sustainable growth, harnessing digital assets’ potential to enhance productivity, and revitalizing the workforce and network of people.”

“Top performing supply chain organizations make investment decisions with a different lens than their lower performing peers,” said Ken Chadwick, VP Analyst in Gartner’s Supply Chain Practice. “Enhancing productivity is the key factor that will drive future success and the key to unlocking that productivity lies in leveraging intangible assets. We see this divide especially in the digital domain where the best organizations are far ahead in optimizing their supply chain data with AI/ML applications to unlock value.”

But what’s the difference between AI and machine learning (ML)? Here, we’ll take a look at some of the distinguishing features and why a new ML model — Optimal Machine Learning (OML) — may hold the key to achieving better supply chain agility and resilience.  

AI and ML

Although various definitions exist, Accenture defines AI as “a constellation of many different technologies working together to enable machines to sense, comprehend, act, and learn with human-like levels of intelligence. …Technologies like machine learning and natural language processing are all part of the AI landscape. Each one is evolving along its own path and, when applied in combination with data, analytics and automation, can help businesses achieve their goals, be it improving customer service or optimizing the supply chain.”

IBM defines machine learning as “a branch of artificial intelligence (AI) and computer science that focuses on using data and algorithms to enable AI to imitate the way that humans learn, gradually improving its accuracy.”

Since AI and ML are so closely related, IBM says it can be “tricky” to try to differentiate between the two: “In fact, machine learning algorithms are a subset of artificial intelligence algorithms — but not the other way around.”

IBM additionally describes the difference by noting that:

  • “AI refers to any of the software and processes that are designed to mimic the way humans think and process information. It includes computer vision, natural language processing, robotics, autonomous vehicle operating systems, and of course, machine learning.”
  • “…machine learning specifically refers to teaching devices to learn information given to a dataset without manual human interference.”

Expounding on the latter and noting that there are “numerous approaches” to ML, IBM says that with this approach to AI, ML algorithms are used that can “learn from data over time in order to improve the accuracy and efficiency of the overall machine learning model.”

Traditional ML models

IBM describes three primary categories of ML models.

Supervised machine learning: “…is defined by its use of labeled datasets to train algorithms to classify data or predict outcomes accurately. As input data is fed into the model, the model adjusts its weights until it has been fitted appropriately. This occurs as part of the cross validation process to ensure that the model avoids overfitting or underfitting. …”

Unsupervised machine learning: “…uses machine learning algorithms to analyze and cluster unlabeled datasets (subsets called clusters). These algorithms discover hidden patterns or data groupings without the need for human intervention. …”

Semi-supervised machine learning: “…offers a happy medium between supervised and unsupervised learning. During training, it uses a smaller labeled data set to guide classification and feature extraction from a larger, unlabeled data set. Semi-supervised learning can solve the problem of not having enough labeled data for a supervised learning algorithm. It also helps if it’s too costly to label enough data.” 

Optimal Machine Learning (OML)

Another type of ML that has recently been described is Optimal Machine Learning (OML). In an article published in the March–April 2024 edition of the Harvard Business Review (HBR), “How Machine Learning Will Transform Supply Chain Management,” Narendra Agrawal, Morris A. Cohen, Rohan Deshpande, and Vinayak Deshpande describe the “shortcomings of traditional planning systems” for supply chains and how OML has “proved effective in a range of industries.”

“A central feature is its decision-support engine that can process a vast amount of historical and current supply-and-demand data, take into account a company’s priorities, and rapidly produce recommendations for ideal production quantities, shipping arrangements, and so on,” according to the article summary. 

OML is a term coined by the team at AD3 Analytics, a startup with which the authors of the article are affiliated in various roles

A September 2021 post describing the company’s beginnings says that its “founding team has had a unique vantage point – witnessing the evolution of supply chain planning over multiple decades – by teaching cutting-edge theory and working with industry. The world has changed significantly, and it is becoming increasingly evident that the capabilities of available supply chain planning solutions has not caught up.”

How OML tackles planning-method pitfalls

In the HBR article, the authors underscore the supply chain disruptions of the past few years and the need for better planning to increase agility and resilience — a challenge that has been an ongoing struggle for many businesses. 

“One major cause is flawed forecasting, which results in delivery delays, inventory levels that are woefully out of sync with demand, and disappointing financial performance,” they write. “Those consequences are hardly surprising. After all, how can inventory and production decisions be made effectively when demand forecasts are widely off?”

To address such a “deficiency,” the authors say their new approach — OML — involves using AI technology “to create a mathematical model that takes key data inputs related to the supply chain” and “links them to planning decisions.”

“This model can take into account a company’s priorities…its budget restrictions, and other resource constraints…,” they write. “The data is stored in a way that enables updating in near real time and quick revision of the calculations that inform decision-making.”

They say the failure to develop “effective strategies” to deal with unexpected supply chain disruptions is a result of “three significant shortcomings” in the planning methods currently in use: 

  • “Flawed, forecast-driven processes”
  • “Data-related challenges”
  • “Ineffective scenario planning”

After providing further details about each, the authors describe the “new paradigm” of OML — which has three key components:

  • Decision-support engine
  • Digital twin
  • End-to-end data architecture

Underscoring how important it is that senior managers “ensure that all parties trust the recommendations that come out of the planning system” without additional review, the authors say that achieving success using OML to build supply chain agility and resilience requires:

  • An “appropriate organizational structure”
  • Personnel with the “right skills”
  • Planning process changes
  • A “detailed understanding of the potential and pitfalls of machine learning”

For additional details about how OML can help improve supply chain agility and resilience, please see the HBR article and/or the AD3 website

Leave a Reply