Harnessing Smart AI: From Disruption Preparedness to Disruption Prevention

The increasing volume and complexity of procurement data are staggering. AI is now necessary for informed, actionable decision-making.

Mr Sdc Exec Native Ads 1600x900

*This article is sponsored by SourceDay*

In recent years, the global supply chain has gained unprecedented attention, extending beyond procurement professionals to capture widespread public interest. However, the issues that have stolen the headlines – geopolitical unrest, logistical bottlenecks, and the recent global CrowdStrike outage that snarled supply chains – are just part of the story. Amidst the huge delays that plagued the transit legs of the supply chain, much more of the risk actually happens before shipment. In fact, 70% of inbound supply chain issues happen prior to shipment, according to Dispatch Track. The cause of these disruptions isn't nearly as dramatic as we might think; mostly chalked up to supplier miscommunication, changes in lead times, prices, or quantities, or simply having a bad email address. In other words, the risk is coming from inside the house!

Understanding the True Origins of Supply Chain Risk

Now that we know where the problem lies, fixing it should be easy enough, right? A little more attention to detail here and some assertive follow-up there, and we should be able to course-correct. Not so fast. In our experience, the level of risk that develops at critical points throughout the PO lifecycle is grossly underestimated. In fact, more than half (52%) of all PO lines require an average of 2.5 manual changes. For example, a medium-sized company might have 300,000 PO lines to manage each year. If half of those require manual intervention, that’s 390,000 updates that require a person to stop what they’re doing and make some kind of manual change. This gets unmanageable fast. The sheer volume of change makes identifying and addressing all of these issues virtually impossible for procurement teams that are already under-resourced.

This means most inbound supply management is inherently riskier than it needs to be. It also means procurement teams are forced to remain reactive, never being able to get ahead of issues. This myopic focus on just keeping things moving doesn’t leave much opportunity to evaluate overall supply chain performance. And the largely manual work opens the door for manual errors and even more risk, compounding an already chaotic situation. One of the foundational obstacles we need to overcome is the attachment the industry has to outdated processes. When your data lives in spreadsheets or email folders, it prevents any real planning or collaboration from happening. 

Supply Chain Management is the Ideal Use Case for AI

Supply chain management is a great example of how AI can make a complex, labor-intensive process more efficient. With the surge in popularity of generative AI tools like ChatGPT, mainstream users are becoming familiar with the fundamentals of how AI works. And while these tools differ from enterprise AI applications, generative AI is a good analogy to consider.  Most of the generative AI tools we hear so much about primarily do three things: predict, generate, and learn. Tools like ChatGPT predict the most logical word to generate in order to produce convincingly authentic text. The effectiveness of the result is reinforced based on user feedback, and then it learns how to make even more convincing versions next time.

These three elements, predict, generate, and learn, are not so different from what happens in supply chain management. Buyers predict what might happen based on the current conditions and where something needs to be, then evaluate the result and apply what was learned to improve the process. This makes AI a natural complement to supply chain management. But there’s a trend right now of companies rushing to incorporate generative AI tools into their platforms, even as the value that generative AI alone can deliver remains nebulous. This gives these companies cover to claim AI as part of their tools without it actually being able to address any of their customers’ problems.

However, even in the face of this ambiguity, one immutable fact remains: AI is only as good as its data set.

Turn Disruption Preparedness into Disruption Prevention

Data is the foundation for SourceDay Intelligence, a new proprietary AI/ML platform that harnesses a decade of expertise to bring solutions-oriented AI to supply chain management. The scale of our data, informed by nearly $60 billion in direct spend through the platform, makes it possible for us to recognize patterns that are inaccessible to anyone else in the industry. We also have insight into every step of the collaboration process. More than 75 million messages between buyers and suppliers provide a unique understanding of the patterns that drive good outcomes versus bad.

Predict, Recommend, Automate

SourceDay Intelligence helps separate the signal from the noise by predicting unforeseen risks, recommending effective resolution, and automating workflows. This makes it possible to focus on more strategic business challenges. It also means procurement teams can be proactive instead of reactive. Rather than waiting to jump on the phone when something’s late, our AI modeling can predict the variables that lead to delays in the first place. By anticipating issues earlier, a wider range of actions can be taken and the issue can be resolved before it causes any disruption. Imagine knowing if something is at risk as soon as the PO is generated. This head start would give you time to collaborate with suppliers and make adjustments before it becomes an issue.

Armed with the data to determine what workflows lead to the best outcomes, SourceDay Intelligence can recommend strategies for improving high-risk areas and automate custom workflows to manage high-frequency and low-impact PO tasks.

The promise of AI has always been to support, enhance, and streamline human decision-making. That approach, coupled with exhaustive data, will have a dramatic impact on the most vulnerable stages of the supply chain.

About Michael Miller

Michael is the Chief Operating Officer at SourceDay and leads Product strategy and management. He has held multiple C-level roles at early stage and emerging supply chain technology companies. Most notably, Michael was COO at fellow Silverton Partners portfolio company Convey Inc for 4 years, overseeing all delivery functions during a period of high growth and through its acquisition by project44.

About SourceDay

SourceDay is a direct materials procurement platform focused on de-risking Purchase Order Lifecycle Management. Powered by patent-pending AI and Machine Learning, SourceDay delivers enhanced visibility, predictability, management, and accuracy to manufacturers and distributors. SourceDay integrates with any ERP system transforming costly, manual, and often error-prone tasks into precision workflows.

Latest