The last decade has taught supply-chain professionals that there are many forces at work in the world that are impacting their organizations. Artificial intelligence and machine learning tools have been shown to be an essential component of the technology portfolio that helps to make sense of all this, and create more responsive and resilient supply chains.
A.I. and machine learning are currently top priorities for organizations that are looking to incorporate technology into their supply-chain planning and execution tasks — and for good reason. They open a window into a world that wasn’t visible before, and can help to automate many repetitive tasks.
Our research finds that there is still little understanding of just what A.I. and machine learning are, and how to deploy them. Even for those who have adopted some of the technology, many have trouble articulating exactly why A.I. is different than what they had been doing before.
Getting Started
Get educated. This is the first step toward taking advantage of A.I. and machine learning. Users need to get grounded in the terminology, then remove the myths and misunderstandings, in order to determine the appropriate starting point and best uses for A.I. in their organizations. Multiple experts are needed to vet terminology that is both accurate and specific to supply-chain purposes.
Get your data in order. The need for users to focus on data inclusion, definitions and quality can’t be emphasized enough A.I. may use existing data in a different manner, while many applications seek out external data sources, so it’s essential to have on hand some good data resource management (DRM) tools that can assist users and provide an expanded view of the operation.
Get the right portfolio of technology. Not just some smart software, but visibility platforms, subscriptions to curated data feeds, and A.I.-powered applications are minimum requirements.
Work with experienced supply-chain providers who have a strong A.I. and machine learning bench. These are challenging times for supply-chain teams. They need to focus on the changes in the world and how their companies can respond to them. It’s important, therefore, to seek out specific expertise in A.I. Experienced providers have invested heavily in either acquiring A.I. companies or recruiting top professionals to populate and develop ready-to-deploy technology and curated data. They should already know the pitfalls of early-stage deployments. Over time, users will encounter broader and more creative challenges than turnkey package can offer.
Revise project-management initiatives and expectations for A.I. and machine learning projects. Users need to understand that with A.I. and machine learning, they are turning on a mega-learning engine. A.I.-based robotic systems have to learn the terrain, and planning systems must access end-to-end supply chain data, in order for the right solution to be coded or selected. The development lifecycle, therefore, has to change. The key to implementing A.I. and machine learning systems is to look at the data first, see the patterns that emerge, then pick the right tools, method and algorithms.
Outlook
In 2021, more A.I. will be deployed across the supply chain, from robotics in warehouses, factories and stores to analytic systems for planning teams. Organizations will focus on new skills training, including key hires such as data scientists. When working with solutions providers, buyers should be assured of their existing capabilities, the experience of the people who will be deploying them, and what the best opportunities for using A.I. will be.
