The AI Replenishment Agent: How It Drafts Purchase Orders in Odoo and MYOB Acumatica

Updated 6 October 2026: added the short answer, an At a glance agent spec, four question headings and a captioned table of what the rule decides, what the agent adds and who signs off; the Odoo section now reflects Odoo 20 (released 24 September 2026), including its suggested minimum and maximum stock levels and what the AI Purchase and AI Stock modules do and don’t contain; the MYOB Acumatica section now cites MYOB’s current AI feature list; the ABS figure was re-sourced to the 25 June 2026 release; “Goes well with” added. The worked example and Bill’s quote are unchanged.

An AI replenishment agent reads sales velocity, stock on hand, supplier lead times and minimum order quantities in Odoo or MYOB Acumatica, then drafts purchase orders that a buyer approves before anything is sent. As at 6 October 2026 neither platform ships that agent natively: Odoo 20 suggests minimum and maximum levels from demand history, and MYOB Acumatica’s replenishment engine is statistical.

At a glance

• Who it is for: buyers, operations managers and owners of Australian wholesale, distribution and importing businesses on Odoo or MYOB Acumatica who carry too much of some lines and too little of others.
• Problem it solves: a reordering rule checks one product against numbers someone typed in months ago. The agent notices when lead times, margins and demand have drifted away from those numbers, and drafts the order a careful buyer would.
• Platforms: Odoo 20 and Odoo 19 (Inventory and Purchase, with the Approvals app on Enterprise) and MYOB Acumatica (replenishment, Distribution Requirements Planning). The agent is built alongside either platform and reads through its API under its own login.
• Trigger: a schedule (Monday 6am in the example below), plus on-demand runs before a supplier’s order cut-off.
• Inputs read: sales orders by product and period, on-hand and incoming stock by location, supplier lead times as quoted and as actually received, minimum order quantities, carton multiples, landed cost and margin.
• What the agent drafts: purchase recommendations grouped by supplier, with quantities that respect minimum order quantities and pack sizes, a short reason per line, and questions where it cannot tell a trend from a one-off.
• Approval gate: the buyer reviews every draft, edits or rejects lines, and only then does a request for quotation or purchase order go out through the ERP’s own approval workflow.
• Audit trail: every read is logged under the agent’s own credentials, every recommendation keeps its reasoning and the records behind it, and the buyer’s edits are kept beside the draft.
• Typical cycle time: the draft is ready before the buyer starts; the review takes about twenty minutes a week in the example below.
• What it never does: confirm or send a purchase order, change a reordering rule, a price or a supplier record, or supply a pack size or lead time it did not fetch from the ERP.
• Typical cost: a scoped pilot from $5,000 ex GST, per the AI agents for ERP in Australia hub as at 6 October 2026; the agent’s login is a paid seat on both platforms.
• Typical timeline: discovery 1 to 2 weeks, sandbox pilot 2 to 6 weeks, drafts-first go-live 1 to 2 weeks, per the same hub page.
• Last verified: 6 October 2026 against the Odoo 20 release notes, Odoo’s 20.0 reordering rules documentation, the Odoo 20 Enterprise source, MYOB’s AI feature page and the ABS Characteristics of Australian Business release of 25 June 2026.

Most wholesale and distribution businesses live with a strange contradiction: too much stock of some products and not enough of others, at the same time, in the same warehouse. The reorder report says one thing, the sales team says another, and the buyer spends Monday morning reconciling both in a spreadsheet. This is the job an AI replenishment agent is built for, and it’s one of the capabilities Australian businesses ask us about most.

The appetite is real. MYOB’s ERP Trends research found 78% of wholesale distribution businesses plan to use AI in their ERP, and the ABS Characteristics of Australian Business release of 25 June 2026 reports that 12% of Australian businesses used AI in 2024-25, compared with 1% in 2022-23. Most of that adoption is still chatbots and document drafting. Replenishment is where it starts touching real money, which is exactly why the design matters.

What does an AI replenishment agent do?

The pattern has four verbs, and only three of them belong to the agent.

• It reads. Sales orders by product and period, current and incoming stock, supplier lead times as quoted and as actually delivered, minimum order quantities (MOQs), carton multiples, landed cost and margin. All of it already lives in your ERP.
• It flags. Lines trending toward a stockout inside the supplier’s lead time. Slow movers tying up cash. Lines where the quoted lead time and the last six receipts no longer agree.
• It drafts. Purchase recommendations grouped by supplier, with quantities that respect MOQs and pack sizes, and a short note explaining each suggestion.
• A person approves. The buyer reviews the draft, edits quantities, rejects lines, and only then does a request for quotation or purchase order go anywhere.

That last step isn’t a training-wheels phase to grow out of. It’s the design.

What do Odoo and MYOB Acumatica already do without an agent?

Here’s the part most AI content skips: both platforms already ship serious replenishment logic, and almost none of it is AI.

Odoo handles this through reordering rules: set a minimum and maximum per product, and Odoo either creates the purchase or manufacturing order automatically or lists the line on the replenishment dashboard for a person to action. Odoo 20, released on 24 September 2026, adds two things per its release notes: minimum and maximum levels are now suggested from demand history, the days of coverage you want and how often you order, and the dashboard’s two order buttons became one. That is a calculation on your own history. It is useful, and it is not an agent.

We’ve been through the Odoo 19 and Odoo 20 Enterprise source code. In the 20.0 source as at 6 October 2026, the AI Purchase and AI Stock modules each add one sample prompt to the AI chat (find the latest purchase order from a vendor; find the most common stock move) and nothing to the replenishment calculation. What Odoo 20 does give you is the parts to build one inside the database: an AI agent can be run by a scheduled action and can hold a skill that creates records, which we covered in our post on Odoo 20 agents that run unattended. Every AI feature in Odoo 20 draws on IAP credits, so a nightly agent over a large catalogue is a budget line as well as a design decision.

MYOB Acumatica, built on the Acumatica platform, goes further on the statistics. Its replenishment tools cover min/max settings, safety stock, seasonality, lead times and demand forecasts, and Distribution Requirements Planning time-phases suggested supply orders across locations. These are statistical models, not machine learning, which is no criticism: they’re predictable, explainable and cheap to run. As at 6 October 2026 MYOB’s AI page lists a built-in AI Assistant that answers plain-language questions about inventory among other things, Anomaly Detection and AI Automation, and names no AI replenishment or demand forecasting feature. For the full list, see our guide to MYOB Acumatica’s AI capabilities.

If you haven’t configured these built-in tools yet, start there, not with an agent. An agent reasoning over wrong lead times and missing MOQs will draft confident nonsense. We covered the groundwork in our guides to Odoo’s purchasing module and Odoo inventory management.

Where does the agent add judgement the rules can’t?

A reordering rule evaluates one product at a time, against parameters someone typed in months ago. It works until the world drifts away from the parameters. The agent’s job is noticing the drift.

Rules can’t tell you that a supplier’s last four receipts averaged twelve days late, so the quoted lead time is fiction and the reorder point is too low. They can’t weigh whether topping up a low-margin line is worth the freight. They won’t spot that you’re simultaneously out of stock in Brisbane and overstocked in Melbourne on the same SKU, where the right move is a transfer, not a purchase. And they can’t read context, like the doubling of sales on three lines that might be a trend or might be one customer’s one-off project.

A worked example: Monday at 6am

Here’s what that looks like in practice. Monday, 6am, the agent runs against live ERP data with read-only access. By 7am the buyer has a draft in their inbox: fourteen lines need attention this week. Nine are routine top-ups, grouped into three supplier orders that respect each supplier’s MOQ. Two are early orders because of that supplier with the drifting lead times, with the receipt history attached as evidence. Three are flagged as questions, not orders, because sales velocity doubled last month and the agent can’t tell whether that’s repeatable, so it presents both order quantities and asks. The buyer approves nine, adjusts one, rejects one, and answers the three questions. Total elapsed time: about twenty minutes, most of it on the three lines that deserved human thought.

“The reorder point tells you one number went below another number. It can’t tell you the supplier has been shipping late for three months, or that the margin on a line no longer covers the freight to restock it. That judgement layer is what the agent drafts, and it’s exactly why a buyer still signs off on every order.”

Bill Alvarez, Practice Manager, Auboros

Why does a buyer approve every order?

Language models make errors. They misread a spreadsheet column, treat a spike as a trend, or hallucinate a pack size. On a chatbot that’s an annoyance. On a purchase order it’s a container of stock you didn’t need, paid for with real cash and stored in real racking.

The approval gate is what makes the agent safe to run at all, and in practice it means three things. The agent’s access is read-only and scoped to the data a buyer could see anyway, never a shared super-user login. Everything it produces lands as a draft, and the ERP’s own workflow handles release. Odoo Enterprise ships an Approvals app that generates the request for quotation from an approved purchase request, so the gate is native to the platform, not bolted on. And every recommendation carries its reasoning and sources, so when the buyer disagrees, you can see why the agent got it wrong and fix the data or the prompt.

If you build the agent inside Odoo 20 with a record-creating skill, the same rule applies with more force: the skill should create a draft request for quotation and an activity for the buyer, and never confirm the order. We set out that pattern in our post on the approval gate for Odoo AI server actions.

This is the same drafts-first pattern we apply across our AI agents for ERP in Australia: agents propose, humans approve, and everything is auditable. Two sibling agents pick up where this one stops: the supplier PO chasing agent confirms delivery dates once the order is placed, and the inventory exception agent handles the stockout-and-excess cases where the answer is a transfer, not a purchase.

How to start without betting the warehouse

Don’t switch this on across your whole catalogue. Pick one category or one supplier group, ideally one with steady demand and a co-operative supplier. Check the data first: real lead times, current MOQs, correct costs. Then run the agent’s recommendations alongside your existing process for a full ordering cycle and measure the gap between what it drafted and what the buyer actually ordered. Where they differ, work out which one was right.

If the agent earns trust, widen the category coverage. If it doesn’t, you’ve lost nothing except a few weeks of parallel running, and you’ve usually found data problems worth fixing anyway. For wholesale distributors starting from scratch, this pairs naturally with a platform foundation like MYOB FastStart for wholesale distribution, which gets the underlying purchasing and inventory data structured properly in the first place.

Goes well with

• The supplier PO chasing agent: what happens after the buyer approves, when the supplier goes quiet on the delivery date.
• The AI demand forecasting agent: how good the forecast underneath the order quantity can be, and how to test it.
• What an AI agent costs to run in Australia: the seats, tokens and review time behind the pilot figure above.

Thinking about a replenishment agent for your warehouse?

We’re a Brisbane-based ERP consultancy that implements Odoo and MYOB Acumatica across Queensland and beyond, and we run drafts-first AI agents over both platforms. If you want to know what a replenishment agent could draft from your own sales and supplier data, book a free consultation. We’ll tell you plainly whether your data is ready for one.