MRPeasy + private AI
AI for MRPeasy Manufacturing ERP
Short answer
AI on MRPeasy means grounding a private model on MRPeasy's own REST API and connected QuickBooks Online or Xero ledger, so planners and operations managers can ask plain questions about production, inventory, and lot traceability and get exception and follow-up agents, instead of exporting MRPeasy's canned reports into a spreadsheet to piece the answer together.
- ERP
- MRPeasy
- Industries
- Manufacturing, Food and Beverage, Electronics Assembly
- Written for
- Operations Manager
MRPeasy is a cloud MRP and light ERP built for small manufacturers who have outgrown spreadsheets: sales and CRM, purchasing, manufacturing with BOMs and routings, inventory, and a visual production schedule, all in one system with a real REST API. It is a common step up for job shops, food and beverage producers, and electronics assemblers in the 5 to 200 employee range.
Because MRPeasy covers more of the business than a pure inventory tool, the questions people ask span modules: whether a work order is behind because of a material shortage, whether a lot of raw material tied to a customer complaint went into other batches, or what a product actually costs once the latest purchase prices are factored in. MRPeasy's own reporting handles the standard version of these; the cross-module, judgment-based version usually does not have a built-in report.
Most MRPeasy customers do not have a dedicated business analyst. The operations manager, plant manager, or owner is answering these questions between other responsibilities, which is exactly the gap a grounded AI layer closes without adding another system for staff to learn.
This page covers what AI on MRPeasy actually looks like: how it connects to MRPeasy's API and accounting sync, what agents can safely do, and where MRPeasy's cloud-only architecture shapes the deployment options.
What usually gets in the way
The problems we hear most from operations manager teams running MRPeasy.
Cross-module questions have no built-in report
Whether a work order delay traces back to a specific material shortage, or what a lot of raw material actually went into, requires joining data across modules that MRPeasy's standard reports do not combine.
No dedicated analyst to build custom views
Small teams rely on whoever is available, usually the operations or plant manager, to manually export and reconcile data for anything outside the standard reports.
Lot and batch traceability lookups are slow
Food and beverage and electronics customers need fast, confident answers on where a lot went for recall or quality investigations; digging through MRPeasy screens manually is too slow under time pressure.
Demand forecasting still needs manual sanity-checking
MRPeasy's forecasting module produces numbers, but explaining why a forecast changed or whether it should be trusted for a specific SKU is a manual, judgment-heavy exercise.
Master data quality quietly erodes trust in reports
Inconsistent BOMs, stale lead times, or incomplete item records mean planners often distrust the system's own numbers and fall back to their own spreadsheets.
Where AI earns its place in MRPeasy
Each use case names the ERP objects it reads or writes, so your ERP team can judge the integration effort before anyone commits budget.
Natural-language production and inventory Q&A
Planners and operations managers ask plain questions about work order status, inventory on hand, and open purchase orders without navigating MRPeasy's menus.
Touches: Manufacturing Orders, BOM, Routing, Inventory, Purchase Orders
Outcome: Cuts routine status lookups from a multi-screen search to a direct answer.
MRP exception triage copilot
An agent surfaces work orders at risk because of component shortages or late purchase orders, ranked by customer due date, rather than the planner scanning the full MRP output.
Touches: MRP run results, Manufacturing Orders, Purchase Orders
Outcome: Turns the daily MRP review from a full-list scan into a short, prioritized action list.
Lot and batch traceability lookup
During a quality issue or recall, the agent quickly answers which finished goods lots contain a specific raw material lot, and which customers received them.
Touches: Lot/batch tracking, BOM, Sales Orders, Shipments
Outcome: Shortens the time to answer a traceability question from hours of manual lookup to minutes.
BOM and cost rollup narrative
The agent explains what changed in a product's standard cost after a purchase price update, referencing the specific components that moved.
Touches: BOM, Routing, Purchase Order pricing history
Outcome: Gives finance and sales a clear reason for a cost change instead of an unexplained number.
Demand forecast explanation
When MRPeasy's forecast for a SKU shifts significantly, the agent summarizes the sales history behind the change so the planner can decide whether to trust or override it.
Touches: Demand forecasting module, Sales Order history
Outcome: Reduces blind trust or blind override of forecast numbers with a grounded explanation.
Purchase order follow-up agent
The agent drafts follow-up messages for late purchase orders, referencing the actual PO lines and supplier history, for a buyer to review and send.
Touches: Purchase Orders, Suppliers
Outcome: Turns manual PO chasing into a reviewed, ready-to-send draft list.
Shift handover summarization
The agent summarizes shop floor notes and work order progress from one shift into a short handover brief for the next.
Touches: Work Order comments/notes, Manufacturing Orders
Outcome: Reduces information lost between shifts without requiring a formal handover report.
Reference architecture
The connector layer pulls from MRPeasy's REST API and the connected QuickBooks Online or Xero ledger on a schedule into a private store, with a semantic layer that maps MRPeasy's Sales, Manufacturing, Inventory, and Purchasing modules to the terms planners actually use.
- 1
MRPeasy and accounting connectors
Read access to the MRPeasy REST API covering sales, purchasing, manufacturing, and inventory, plus a read-only view of the connected QuickBooks Online or Xero ledger.
- 2
Semantic layer
Resolves MRPeasy's module and field names to planner-facing concepts (available to promise, at-risk work order, lot genealogy) consistently across every question.
- 3
Model serving
An open-weight model sized for a small manufacturer's data volumes, served on infrastructure the customer controls rather than a shared public endpoint.
- 4
Retrieval and agents
Grounded question answering plus narrow agents (MRP exception triage, PO follow-up drafting) that stop at a recommendation or draft for a human to approve.
- 5
Governance and audit
Every answer traces to the specific MRPeasy records it used, and lot traceability answers are logged for quality and recall documentation purposes.
Integration notes for your ERP team
- MRPeasy provides a documented REST API covering sales, purchasing, manufacturing, and inventory objects; this is the integration path, since MRPeasy has no customer-accessible database.
- The QuickBooks Online or Xero connection should use a separate, read-only, scoped credential from MRPeasy's own accounting sync to avoid any risk to that sync.
- MRPeasy's lot/batch tracking fields need explicit mapping in the semantic layer, since traceability questions during a quality event have to be fast and unambiguous.
- MRPeasy has no on-prem edition, so 'on-prem AI' means the AI infrastructure runs on-prem or in a private VPC while MRPeasy stays cloud-hosted; state this plainly to the customer.
- The demand forecasting module's outputs are accessible via API and are a useful grounding source for forecast-explanation use cases, rather than something to recompute independently.
- MRPeasy's CRM module often holds early customer commitment data that is useful for prioritizing MRP exceptions by actual customer importance, not just due date.
Deployment options
Private cloud AI layer
The default for MRPeasy customers, since MRPeasy has no on-prem edition
Connectors, data store, and model run in a private VPC the customer controls; MRPeasy and the connected accounting platform remain cloud SaaS.
Air-gapped data layer for regulated cases
Food and beverage manufacturers with recall traceability obligations, or electronics assemblers with customer IP or export-control constraints
Scheduled pulls mirror MRPeasy data into an isolated on-prem environment where lot traceability and cost queries run without any outbound network path.
Hybrid
Multi-site manufacturers running a shared MRPeasy instance across several facilities
A central private model serves every site from the same consolidated data store, while MRPeasy itself stays the single cloud system of record.
Compliance and data control
How the architecture supports your obligations. Certification and accountability stay with your organisation; the design keeps the evidence straightforward.
Food safety and traceability (FSMA/HACCP context)
Lot traceability answers are grounded on the actual MRPeasy lot and BOM records and logged, supporting the documentation trail a recall or audit requires.
Data residency
Since MRPeasy is hosted in the EU (Estonia) with regional options, the private AI layer can be deployed in whichever region satisfies the customer's own residency requirement, independent of MRPeasy's hosting.
GDPR-aligned data minimization
Customer and employee data pulled for grounding is limited to what use cases need; broader personal data stays out of the model's context.
Export control (where relevant)
Electronics assemblers on MRPeasy doing export-controlled work can use the air-gapped pattern to keep BOM and customer data off shared infrastructure.
Where Netray fits
ERPray
Grounded natural-language question answering over MRPeasy's production, inventory, and purchasing data, including lot traceability, fits ERPray's model directly.
Custom build
Specific agents such as MRP exception triage tuned to a company's own due-date and customer-priority rules typically start as a custom build on top of the same connectors.
How an engagement runs
Phase 1 . 1-2 weeks
Discovery
- -Review of MRPeasy module usage, lot tracking configuration, and connected accounting platform
- -Inventory of the questions currently answered by manual export and reconciliation
- -Assessment of any food safety or export-control traceability requirements
Phase 2 . 3-5 weeks
Pilot
- -Connectors and semantic layer covering production, inventory, purchasing, and lot tracking
- -Natural-language question answering validated against real historical questions
- -One agent (MRP exception triage or PO follow-up) running in review-before-action mode
Phase 3 . 2-3 weeks
Production
- -Private-cloud or air-gapped deployment matched to the customer's regulatory needs
- -Access controls aligned with who currently sees cost and customer data in MRPeasy
- -Handover so the operations manager can operate it without ongoing vendor involvement
Phase 4 . ongoing
Scale
- -Additional agents added as new repetitive manual tasks are identified
- -Coverage extended to additional facilities on a shared MRPeasy instance
- -Periodic review of forecast-explanation and exception-triage accuracy
Questions to ask any vendor, including us
A short list that separates real MRPeasy AI work from a chatbot demo.
- Does the AI layer write back into MRPeasy or the connected accounting platform, or is it read-only?
- Since MRPeasy has no on-prem edition, where does the AI infrastructure run and who controls it?
- How are lot traceability answers logged for recall or audit purposes?
- Is customer and employee data minimized to what each use case actually needs?
- How is an MRP exception or PO follow-up recommendation reviewed before anything is sent?
- Can the deployment region be chosen independently of where MRPeasy itself is hosted?
- What happens as our BOMs, routings, or lot tracking configuration change over time?
Frequently asked questions
Can MRPeasy AI stay private if MRPeasy itself is cloud-hosted?
Yes. MRPeasy remains cloud SaaS, but the AI layer, including the data store, model, and agents, runs in infrastructure the customer controls, and no MRPeasy or accounting data is sent to a public model API. That achieves data control without requiring MRPeasy itself to change how it is hosted.
Does MRPeasy have built-in AI already?
MRPeasy's core strength is production planning, BOM/routing management, and integrated purchasing and inventory, not a native generative AI layer. A private AI layer adds natural-language question answering and agents on top, grounded in MRPeasy's own data.
How fast can a lot traceability question be answered?
Once the semantic layer maps MRPeasy's lot and BOM records correctly, a genealogy question, such as which finished lots contain a specific raw material lot, resolves in seconds rather than the manual multi-screen lookup it usually takes today.
Is this useful for a food and beverage manufacturer specifically?
Yes. Food and beverage MRPeasy customers get the most direct value from fast, logged traceability answers during a quality investigation or recall, which is exactly the kind of judgment-heavy, cross-module question MRPeasy's standard reports do not cover.
Can the AI explain why a demand forecast changed?
Yes. The agent summarizes the underlying sales history and any anomalies behind a forecast shift for a given SKU, so a planner can decide whether to trust or override MRPeasy's forecast, rather than treating the number as a black box.
Does this replace MRPeasy's own reporting?
No. MRPeasy's standard reports remain the system of record for routine operational views. The AI layer handles the cross-module, judgment-based, or ad hoc questions that fall outside those standard reports.
How long does an MRPeasy AI pilot usually take?
A working pilot with natural-language question answering and one agent is typically achievable in three to five weeks, since MRPeasy's documented REST API avoids the slower custom integration work legacy MRP or ERP systems require.
Related guides
Natural Language Query for ERP Data: Ask SAP, Infor, or Oracle a Question in Plain English
See how natural language query over SAP, Infor, Oracle, and NetSuite data works: grounded text-to-SQL, role-based permissions, and a visible audit trail.
RAG + SQL + permissionsA Private LLM Grounded on Your ERP Data
How a private LLM answers questions on your ERP data: RAG plus text-to-SQL, role-based permissions inherited from the ERP, and where each fits.
Agents + approval gatesAI Agents for ERP, Running On-Prem
A practical guide to on-prem AI agents for ERP: what they can safely automate, where human approval belongs, and how to design the guardrails.
ERP AI Cost GuideWhat ERP AI Actually Costs: A CFO's Guide
A CFO's guide to what ERP AI actually costs: GPU hardware, model licensing, integration and connector work, and realistic ongoing run-rate ranges.
QAD + on-prem AIAI for QAD Adaptive ERP in automotive and industrial manufacturing
Add AI to QAD Adaptive ERP or Enterprise Edition for automotive and industrial manufacturing, grounded on QXtend and QAD's API layer, on-prem or private cloud.
Plex + private AIAI for Plex Manufacturing Cloud, Grounded in Your Production Data
Put AI on Plex production, quality, and genealogy data without a public model. On-prem or private-cloud LLMs, RAG, and agents built for Plex shops.
Plan it with numbers
Job Shop ERP Fit Assessment
Score how well your ERP handles high-mix, low-volume and make-to-order manufacturing across quoting, one-off routings, WIP visibility, subcontract tracking, and job-level cost variance.
Free ToolNatural Language ERP Reporting Savings Calculator
Turn report volume, build time, and automation rate into the monthly hours and dollars saved by letting users ask ERP questions in plain English instead of building reports.
Free ToolERP Data Quality for AI Assessment
Score item master, customer, vendor, and transaction data quality to find out whether your ERP is ready to ground an AI copilot, forecast, or chatbot.
GuideAI Strategy for the Mid-Market Manufacturer
AI strategy for the mid-market manufacturer: how $20M-$500M firms compete with AI without enterprise budgets, staff, or multi-year transformations.
GuideAI Inventory Optimization for Manufacturers
AI inventory optimization for manufacturers: cut stock 15-30% while raising service levels, with ML-driven safety stock and reorder policies inside your ERP.
GuideYour First AI Agent: A Manufacturing Playbook
Your first AI agent in manufacturing: pick a bounded ERP use case, scope data access, set guardrails, and ship a working pilot in 6 to 8 weeks.
Talk it through with an engineer who knows MRPeasy
Bring one real question your team cannot answer from the ERP today. We will map the data path, the model, and where it runs, and tell you honestly if AI is the wrong tool for it.