5 Ways Predictive AI Can Keep CG Supply Chains Flowing

a view of a large city

I’ve been working many years leading supply chain operations and advising global brands. There’s an inescapable fact that is clear to me: The planning systems we have used for the last 20 years have a huge gap. They fail to “see” a huge number of exceptions that will occur in the next 12 weeks. 

Based on our research, this blind spot can represent 70% of the total exceptions (e.g., stock outs, excess, etc.). The implication  our teams are not working on the right proactive planning actions much of the time.

With Covid-19, the planning gap grew to unmanageable proportions and in some cases, planning models were turned off in favor of manual planning. Still, planners persevered; it was tough, but they were able to recover and keep goods flowing to meet the demand.  Unfortunately, this was tough on planning teams and not sustainable. There is now a capability with AI to truly augment planning teams  making them dramatically more effective.  

Artificial Intelligence (AI): A Force Multiplier for Supply Chain Decision-Making

Advances in computing, data storage and machine learning support planning capabilities that were not possible five years ago.  Powerful capabilities ingest massive data sets, find relationships, predict future outcomes, and recommend new actions that planners understand.  Planners shift from a work day primarily focused on cutting up data to find exceptions and validate information, to one where they have a precise understanding of issues that need a human planner to problem solve.

Kellogg’s, for example, was looking to gain even deeper visibility in their supply chain operations and make better decisions to maximize fill rate and right-size inventory, which is non-trivial when dealing with dated product. For that purpose, they implemented AI inventory and production planning products on top of their ERP to help the planning team understand the dynamics of their business and make decisions improving the flow of products.

According to George Chumakov, Kellogg’s VP of customer service and logistics, AI was a game-changer: “AI takes junior planners and turns them into advanced planners at a rapid rate. It gives everyone the insight that a 20-year employee has been able to achieve.”

The Strategic Application of Predictive AI in Supply Chain Operations

Here’s a quick look at five ways companies are already using AI to create flow, manage volatility and reduce waste:

1. Predicting supply-demand imbalances across the network: Unfilled orders are lost revenue. AI helps companies see demand-side predictions and adjust production accordingly, so they don’t get saddled with excess inventory that undermines their profitability. They can prioritize risks based on business impact and get recommendations for optimal production schedule trade-offs.

2. Optimizing inventory management: Supply chain planners supported by AI can predict supply-demand gaps within the critical 0-to-12-week execution window. They’re free from the fixed rules, data blind spots and backward-looking data found in enterprise resource planning (ERP), advanced planning, and business intelligence systems. Here again, AI can provide insight on how to prioritize risks. It can also recommend optimal actions for improving fill rates at lower costs.

When companies use AI to bring together all their forecasting inputs — machine and human — across the company, it’s a transformative process.

3. Augmenting demand planning and forecasting: Demand and supply plans are notoriously error prone. That’s because consumer goods supply chains rely on too many manual processes and lack a single source of truth for timely, accurate decision-making.When companies use AI to bring together all their forecasting inputs — machine and human — across the company, it’s a transformative process. It enables a demand signal that’s driven by data and backed by a consensus. That, in turn, reduces bias in the network and turns exceptions into rarities.

4. Identifying and unlocking VAR: Value-at-risk (VAR) is a powerful metric. It provides a well-defined picture of the total financial risk facing a brand, region or business unit, and it supports more effective supply chain planning. With AI, companies can unlock the risks previously buried in their supply chain data — dynamic, granular risks they’ve never been able to see before — and collaborate on how to mitigate them before they create disruption.

5. Improving product quality: With help from AI, supply chain operators can unearth previously hidden patterns in specifications variability and defects in product dimensions, composition and more, so they can predict their occurrence. AI can also recommend actions for preventing issues and help teams prioritize those actions based on the cost to the business. That means more quality products entering the market, and subsequently more reliable supply for the front-line professionals responsible for manufacturing and distributing products

Build or Buy: A Word of Caution

The examples above show that AI isn’t just another technology Band-Aid. It’s a solution that makes supply chains more resilient and less wasteful.

That said, I wouldn’t recommend CG companies build these use cases in-house, for the same reasons that you would not build an ERP or CRM from scratch. The level of effort is beyond what nearly any company would consider. 

Should you get dedicated teams of data scientists that begin to work across your company with new tool sets and capabilities — definitely.  Just make decisions about what you build and what you buy with realistic expectations. 

There will always be complexity and volatility in the supply chain. That is inevitable. But by bringing AI into the CG supply chain now, companies will be better prepared to respond to future disruptions, big and small, and keep goods flowing.

Mike Hulbert is vice president of consumer business at