Calculating Planning Parameters Is Just the Beginning
This is the second article in my series on AI-driven planning transformation. In the first article, I discussed why planning parameters drift over time—and how AI can continuously keep them aligned with reality. Updating planning parameters is often one of the fastest AI wins a supply chain organization can achieve. But it isn't the finish line. It's only the first step in building an inventory strategy.
One of the biggest misconceptions I see is that inventory optimization is complete once the planning parameters have been calculated.
It isn't.
In fact, that's where the real work begins.
Whether you're calculating safety stock, reorder points, min-max levels, or lot sizes, the output is simply a mathematical recommendation based on historical performance and statistical assumptions.
Those calculations are important. They provide better inputs to your ERP system and often produce quick improvements.
But they are not an inventory strategy.
An effective inventory strategy has four distinct phases:
1. Calculate – Generate planning parameters using data, analytics, and AI.
2. Analyze – Apply business judgment to determine the right inventory policies.
3. Align – Build agreement across Finance, Commercial, Operations, Purchasing, and Quality.
4. Execute & Monitor – Roll out the new policies, communicate why they changed, and continuously monitor performance.
Most companies spend nearly all of their effort on the first phase.
The organizations that consistently improve inventory performance understand that the remaining three phases determine whether those calculations actually create business value.
This article focuses on Phase Two: Analyze.
The Formulas Don't Know Your Business
Inventory formulas answer a very specific question:
"Based on the data, what should the planning parameter be?"
What they don't understand is your business.
The same mathematical recommendation can be exactly right for one company and completely wrong for another.
Consider three examples.
An industrial manufacturer may carry months of inventory for a five-cent component from Asia because a shortage could stop a production line worth millions of dollars.
A low-margin distributor may intentionally allow slower-moving products to backorder because inventory carrying costs outweigh the lost sales.
A service organization may experience highly intermittent demand because equipment failures can't be forecast accurately.
The mathematics may be similar.
The business strategy is not.
Planning parameters also can't answer questions like:
Is this customer strategically important?
Are we intentionally growing or shrinking this product line?
Can the business afford additional working capital?
Do quality concerns justify carrying additional inventory?
Is supplier performance becoming more variable?
Are geopolitical or logistics risks increasing?
Those aren't mathematical questions.
They're management decisions.
The formula provides precision.
Business leaders provide judgment.
AI Doesn't Replace Judgment—It Makes Better Judgment Possible
For years, inventory planners spent much of their time maintaining planning data.
Updating lead times.
Reviewing safety stock.
Maintaining reorder points.
Cleaning spreadsheets.
There was very little time left to actually analyze the business.
AI changes that equation.
Instead of manually reviewing thousands of SKUs, AI can continuously evaluate demand patterns, supplier performance, lead-time variability, inventory behavior, and service performance.
It can identify where assumptions have changed.
It can recommend which items deserve attention.
It can even suggest planning policies based on the business rules your organization has established.
The planner's role shifts from maintaining data to making better business decisions.
That's where human expertise creates value.
Analysis Starts with Segmentation
Good decisions require structure.
Otherwise, every planner manages inventory differently.
One of the most effective approaches I've seen combines traditional ABC analysis with XYZ demand segmentation.
ABC tells you how financially important an item is.
XYZ tells you how predictable demand is.
Together they create a much richer picture of your inventory portfolio.
X items have stable, predictable demand.
Y items have moderate variability.
Z items experience erratic or intermittent demand.
When these dimensions are combined, different planning strategies naturally emerge.
A Real Client Example
One industrial manufacturer we worked with struggled with poor forecast accuracy, recurring shortages, and growing inventory.
Leadership wanted better forecasting.
What they actually needed was better segmentation.
After analyzing the portfolio, we found something surprising.
Nearly 40% of the company's inventory investment was concentrated in only about 3% of its items.
Those were primarily AZ items—high-value parts with highly unpredictable demand.
Treating those items the same way as stable AX inventory guaranteed shortages and unnecessary expediting.
The solution wasn't to ask planners to spend more time reviewing every SKU.
It was to spend more time on the right SKUs.
The AZ items received customized forecasting approaches, different supplier relationships, and tighter planner oversight.
Meanwhile, the AX inventory was largely automated and continuously monitored by AI.
Lower-value CX items, representing the majority of SKUs but only a small percentage of inventory value, were managed using automated min-max replenishment with AI monitoring for significant changes.
Instead of asking planners to manage everything equally, the company aligned planner attention with business risk.
That is what analysis should accomplish.
Analysis Creates Inventory Policy
The purpose of analysis isn't to create another set of numbers.
Its purpose is to define how the business intends to manage inventory.
For every inventory segment, organizations should answer questions like:
What service level are we trying to achieve?
How much inventory risk are we willing to accept?
Which replenishment method fits this category?
How much planner attention does this segment deserve?
Under what circumstances should we override the mathematical recommendation?
Those decisions transform planning parameters into inventory policy.
They also create competitive advantage because they reflect the realities of your customers, suppliers, operations, and financial objectives—not simply statistical averages.
AI Makes Analysis Continuous
Analysis isn't something that should happen once a year.
Demand changes.
Suppliers improve—or deteriorate.
Products move through their life cycle.
Customer priorities shift.
An item that belonged in one planning segment twelve months ago may require a completely different inventory policy today.
The real advantage of AI isn't that it performs calculations faster.
It's that it never stops watching.
Every day, AI can compare actual demand, supplier performance, and inventory behavior against your planning assumptions.
When something changes, it alerts planners before those assumptions begin driving poor inventory decisions.
That's how inventory management evolves from a periodic exercise into a continuous capability.
Looking Ahead
Even the best inventory policy won't succeed unless the rest of the organization supports it.
In the next article, I'll discuss Align—how Finance, Commercial, Operations, Purchasing, and Quality evaluate inventory policies together, understand the tradeoffs, and agree on a strategy before new planning parameters are deployed.
Then we'll finish the series with Execute & Monitor—how organizations roll out new planning policies, communicate the reasons behind the changes, measure results, and use AI to continuously monitor for drift so inventory strategies remain aligned with changing business conditions.
Because calculating better planning parameters isn't the goal.
Building an inventory strategy the entire organization understands, supports, and continuously improves—that's where the real value is created.
AI Disclosure
Writing & Industry Experience: Greg Pitstick
Editing: ChatGPT
Image: Google Gemini