With one or two exceptions, we have studied an application of artificial intelligence and machine learning in an area of supply chain every week since that first column, with an emphasis on how the innovation highlighted in each article solves customer problems.

That being said, there is also the danger that insufficient sophistication will leave customers at risk of wasting time and money in understanding the basics of AI and machine learning.

This leads to the conclusion that before an AI and machine learning initiative or product is implemented by a company or any organization in its operations, it must first run at least one small and relatively inexpensive pilot in an environment that enables it to assess the reliability and accuracy of the product it considers in relation to the status quo.

From the conversations I have had, it seems obvious that, for certain forms of AI and machine learning, the availability of data of sufficient quantity and quality for decision-making remains a major stumbling block to the application of AI and machine learning to solve most supply chain problems; old data is messy and stored in non-uniform formats in silos.

That old data is also needed for training models and algorithms in AI and machine learning.

Companies with a history of treating data as a strategic asset are better placed to embrace AI and machine learning than their peers who have not.

I have come to the conclusion that AI and machine learning startups that develop proprietary and scalable approaches for their customers to solve this problem may have an edge beyond their peers.

Companies have to select the AI and machine learning startups they are partnering with to match the scope of the problem they are trying to resolve.

There is no shortage of opportunities to apply AI and machine learning in global supply chains to solving big problems.

Comment: Understand the data issues that slow Industrial AI adoption.

Comment: How AI and machine learning enhance visibility in the supply chain, shipping insurance.

Machine learning is streamlining freight forwarding workflows, customs brokering.

Comment: Can AI and machine learning make the economy even better?

Comment: How Japan's ABEJA helps big business to operationalize AI, machine learning.

Comment: Pathmind applies AI, to industrial operations machine learning.

Comment: The Demand Chain applies AI, machine learning to profitability in the retail supply chain.

Machine learning for Gro Intelligence generates insights on the global ag.

Read the original article "Commentary: 3 lessons learned about supply chain AI, machine learning - FreightWaves" at https://www.freightwaves.com/news/commentary-3-lessons-learned-about-supply-chain-ai-machine-learning