AI supply-chain optimization · Explainable by design
Forecasts you can defend.
Optimization you can trust.
Echo Optima turns demand, inventory, and supply decisions into recommendations your team can actually stand behind. Every forecast and every per-SKU reorder recommendation shows its drivers, so you can defend the number in the S&OP meeting instead of guessing at it. The power of an enterprise planning suite, without the seven-figure invoice or the 18-month rollout.
No black box. No hardcoded "accuracy" badge. Every number shows its work.
Built for supply-chain and S&OP teams at multi-node mid-market enterprises. The ones stuck between brittle spreadsheets and seven-figure suites.
Agentic AI · The replenishment agent · Early access
From recommendations you apply to an agent that proposes the whole replenishment move.
Echo Optima is adding an agentic layer on top of the forecast-to-reorder loop. The replenishment agent watches demand drift and stockout risk across your SKUs and nodes, and when something moves, it does four things.
01
Detects
Which SKUs need a reorder-point change or a purchase order.
02
Proposes
The action with its rationale: the demand signal, the safety-stock and reorder math, and the supplier terms behind it.
03
Acts within the limits you set
Small, low-risk changes can apply automatically. Anything above your thresholds is queued for a planner to approve.
04
Records everything
Every applied change, agent or human, lives in the same audit trail.
Agentic, but still defensible.
This extends the core promise. It does not break it.
Act-with-approval by default
The agent proposes. A planner approves consequential moves.
Explainable
Every proposed action carries its drivers, so you can defend it in S&OP.
Guardrailed
You set value and quantity caps and decide which actions may auto-apply. The agent fails closed.
Audited
Every applied change, agent or human, is traceable.
Status: early access
In design-partner validation. We are not claiming a self-driving supply chain. We are proving an agent that does the legwork and asks before it acts.
The problem
You are running a multi-node network on a stack that cannot keep up.
Your ERP is a rear-view mirror. It tells you what already happened. Everything forward-looking, how much to hold, when to reorder, which supplier to lean on, gets re-derived by hand every cycle, in spreadsheets, leaning on one senior planner’s gut. Both of your alternatives fail you in a different way.
Spreadsheets leave you exposed
One statistical method, no demand sensing, no explanation. The forecast becomes one person’s intuition, which means it cannot scale, cannot be audited, and walks out the door when that person does. Safety-stock and reorder math get re-keyed every cycle, slowly, with mistakes.
Enterprise suites hand you a black box
Kinaxis, o9, and Blue Yonder run heavier models, but they still hand you a number with no reasoning you can interrogate. After a 9-to-18-month implementation and a seven-figure commitment you were never sized for.
And being wrong compounds. Too much inventory ties up working capital. Too little causes stockouts. A reorder point set by feel quietly bleeds cash in both directions, every cycle, across every SKU and every node.
You do not have a forecasting problem. You have a defensibility problem. The number is not the hard part. Defending it to your VP and your CFO is.
How it works · the forecast-to-reorder loop
From a demand signal to a reorder you can defend, in one loop.
This is the wedge. Echo connects the forecast directly to the action it should trigger, with the reasoning carried all the way through.
01
Forecast, and explain it.
Holt double-exponential smoothing sets a transparent statistical baseline, level and trend, no hidden weights. Claude AI then layers in qualitative signals, flags what is shifting, and explains the move in plain language. You get trend direction, ranked alerts so you work the biggest shifts first, and the why behind each one.
02
Translate the forecast into a reorder recommendation.
That demand signal flows straight into per-SKU policy. Safety stock and reorder point, calculated for each SKU at each node, with the drivers shown. No re-keying into a spreadsheet. No separate tool. The forecast and the reorder live in the same loop.
03
Review, edit, and apply, with an audit trail.
Recommendations are human-in-the-loop by design. A planner reviews each one, edits it if real-world context demands, and applies it. Every applied change is captured in an audit trail, so the decision and who made it can be traced later.
04
Defend it.
Because every step carries its reasoning, the output is something you can paste straight into your S&OP deck and stand behind. Not "the system said so." The driver, the math, and the call, attached.
Key capabilities
One platform for the decisions that move money.
Everything below is built today.
Explainable demand forecasting
A hybrid of Holt double-exponential smoothing (the auditable statistical baseline) and Claude AI (qualitative sensing and plain-language reasoning). Ranked alerts, trend direction, and the driver behind every forecast.
Per-SKU reorder recommendations
Safety stock and reorder point, calculated per SKU. Editable before you apply. Every applied change recorded in an audit trail. This is the heart of the forecast-to-reorder loop.
Inventory optimization
Right-size what you hold across the network instead of carrying excess as insurance against your own forecast.
Supplier and contract management
Supplier reliability and contract terms feed directly into reorder and safety-stock policy. The recommendation reflects who you actually buy from and on what terms.
Capacity planning
Detect bottlenecks across labor, equipment, and facility before they break service levels.
Network optimization
Compare baseline against an optimized design, with concrete proposed actions (keep, expand, consolidate, open new, or close) for each node.
Modeled across your real network: Plant → DC → FC → Store.
The explainability differentiator
Defend every number in the S&OP meeting.
Incumbent suites give you a number from a heavy engine you cannot open. Spreadsheets give you a number you typed yourself. Echo gives you a forecast and the argument behind it.
That is not a tagline. It is the core design choice. Every forecast surfaces its drivers. Every reorder recommendation shows the safety-stock and reorder-point logic, the supplier terms behind it, and the demand signal that triggered it. When your VP asks "why is this number higher than last quarter?", you have an answer that is not "the model said so."
We say "auditable" and "act-with-approval," not "autonomous." We do not claim a self-driving supply chain. We are adding an agentic layer (the replenishment agent, in early access), but it proposes actions a human reviews, edits, applies, and can defend. It auto-applies only small, low-risk changes within limits you set, with an audit trail behind every applied change. That is the honest, useful middle the mid-market actually needs.
A note on honesty
We used to show a "94% forecast accuracy" badge. It was a hardcoded number, not a measured result, so we removed it. Real backtested accuracy varies widely by SKU and by how clean your history is. Any vendor showing you a single universal accuracy figure is showing you marketing, not math. We would rather show you the drivers and let you judge.
Who it is for
Built for the team spreadsheets outgrew and suites priced out.
Echo Optima is for supply-chain and S&OP leaders at multi-node mid-market enterprises (multiple plants, DCs, fulfillment centers, sometimes stores) who have outgrown spreadsheets but cannot justify a seven-figure planning suite or the transformation program that comes with it.
The demand planner / S&OP manager
Gets blamed for both stockouts and excess, and spends more time assembling the analysis than deciding on it.
The VP of supply chain / COO
Owns working capital and fill rate. Needs decisions that hold up under scrutiny.
The procurement lead
Wants reorder and safety-stock policy that actually reflects supplier reliability and contract terms.
If you have been priced out of Kinaxis and outgrown by Excel, this is the third option.
Integrations
Works with the systems you already run. Honestly.
Echo reads from your system of record so it can deliver the forward-looking forecasting and optimization those backward-looking systems do not. It connects through open APIs and a connector layer.
Available today
API-based data ingestion for demand history and SKU/node data.
On the roadmap
Prebuilt ERP/WMS connectors (SAP, Oracle, and others) and additional source-system integrations.
We will tell you exactly what is live versus roadmap for your stack on the demo call. We will not show you a wall of logos and let you assume they are all one-click today.
Social proof
Validated proof, coming as our design partners go live.
Echo Optima is onboarding design partners now. Customer logos, named case studies, and specific validated metrics will appear here only once a customer or design partner has confirmed them and consented to be referenced, with the source noted.
Echo is built to move the metrics your CFO cares about: working capital, fill rate / OTIF, stockout rate, and carrying cost. Validated numbers will replace category framing as soon as design partners give us permission to publish them.
See it on your scenario
Stop defending a number you do not trust.
See Echo forecast your demand, explain its reasoning, and turn it into a per-SKU reorder recommendation you can defend, in a 30-minute working demo using your own scenario.
Forecasts you can defend. Optimization you can trust.
Need this tailored to your environment?
Every Aelix product can be configured, extended, or built bespoke for your industry, data sources, and compliance constraints. Talk to our engineers about what would change.
Configurable workflows
Adapt rules, thresholds, and approval flows to match your operational policies.
Custom data integrations
Connect to your specific ERP, MES, SCADA, CRM, or proprietary systems.
Bespoke modules
Build product extensions tailored to your industry, region, or compliance needs.