From Experimentation to Transformation: Generative AI in the Supply Chain

From Experimentation to Transformation: Generative AI in the Supply Chain

For most organizations, GenAI is still in the experimental phase, and business leaders may be unsure of how to identify the best use cases within their organizations to leverage as a jumping-off point.

Supply chain use cases can generally be categorized into one of three tiers:

1. Improvements to existing processes
2. Reimagining differentiated processes
3. Disruptions that make significant changes to the top line

With each tier, the impact on the organization, its customers, and its suppliers increases. Let’s identify a few examples of each:

Improvements. Process improvements look for ways to do current work more effectively or more efficiently. Organizations can minimize manual invoice review and processing, or leverage GenAI to prepare the sourcing team for negotiations with shippers.

Reimagination. In this tier, use cases become less focused on manual processes or standardizations and more on heart-of-the-business areas. An organization could use GenAI to generate metadata based on image analysis on a production line for quality control, or to provide insights into monthly sales and operations processing runs that blend news stories and hyper-localized weather forecasts with ERP and planning software insights.

Internally, GenAI can generate personalized micro-training programs for shop floor operators based on insights into shop floor quality issues. During the reimagining process, it’s important for business leaders to identify all potential opportunities, as they are typically more non-linear and require deeper insights than IT alone can identify.

Disruptions. At this tier, the use cases could start to sound like something from a science fiction novel. Clinical trials are already starting to incorporate smartwatch data to capture biometric inputs, speeding the identification of side effects for new drugs.

Instead of fast fashion trailing the runways weeks or months later, social media trends could be used to analyze streetwear, capturing cutting-edge trends, and beating haute couture to the punch.

To accelerate your organization’s transition from “executive education sessions” to an executable roadmap, the most advanced companies identify use cases in the first and second levels as opportunities for both IT and the business to learn. Moving from proof of concept to pilot to full enterprise adoption requires an integrated approach across people, process, data, and technology to address the challenges around scaling and adoption.

People. Understand the change impacts on your workforce and proactively address each with communication, training, and empathy.

Process. Understand the up- and down-stream impacts of any process changes as well as regulatory or supplier impacts.

Data. Understand the sources, including potential risks, when integrating increasing amounts of open-sourced information.

Tech. Understand the pros and cons of developing use cases as an early adopter vs. waiting for standardized use cases to be incorporated into off-the-shelf software and major ERP packages.

If organizations don’t proceed forward with thoughtfulness and care, GenAI experiments will fail long before they ever have a chance to progress to reimagination, or the true end goal: disruption of existing processes to create a brave new world.