MAPICS and Infor XA Private AI Layer Blueprint Generator
This free generator drafts a private AI architecture plan for MAPICS and Infor XA on IBM i. It covers the DB2 for i data path, change capture, off-box model serving and authority mapping, and exports to PDF or Word.
Fill in
MAPICS and Infor XA Private AI Layer Blueprint Generator
Private AI Layer Architecture Plan: MAPICS and Infor XA
Company: Lakeshore Industrial Corp System and version: MAPICS and Infor XA, Infor XA 7.x Database: DB2 for i on IBM i 7.4 Hosting: On-prem server Named users: 250 Sites or plants: 3 GPU: 1-2 x 48 GB class GPUs Target go-live: 2027-06-30
Context and objectives
Lakeshore Industrial Corp wants an AI assistant over its MAPICS and Infor XA data and documents that runs inside its own environment (On-prem server). The assistant must answer questions in natural language, show the data behind each answer, respect existing user permissions and remain read-only until approvals are in place.
Main use cases: MRP exception questions for planners. Purchase order status for buyers. Item and customer order lookups for customer service.
Data boundaries
The following data must never leave the site and must stay out of any external service: Cost rollups and margins. Customer contract pricing. Controlled technical data.
All processing, indexing and model inference run on infrastructure controlled by Lakeshore Industrial Corp.
Components
| Layer | Design | Notes for Lakeshore Industrial Corp |
|---|---|---|
| Data access path | Journal-fed or scheduled replica of approved XA files on x86, with views and a glossary of terse XA field names, reached over JDBC or ODBC | File names are release specific; discover them via the DB2 for i catalog on DB2 for i on IBM i 7.4 |
| Retrieval index | Vector index of drawings, POs and quality records tagged with XA item or order numbers | Built off-box from the IBM i |
| Model serving | Open-weight model on vLLM or Ollama on an x86 GPU server sized to 1-2 x 48 GB class GPUs | IBM i does not run inference |
| Tool server | Typed read tools over the replica (MRP exceptions, PO status, item availability) | No free SQL against base files |
| Identity and permissions | Least-authority IBM i service profile, row and column access control where used, XA roles mirrored in the tool layer | Watch adopted authority paths |
| Audit | Append-only log plus IBM i journal of service-profile access | Reviewed with the IBM i security administrator |
Security controls
| Control | Implementation | Driven by |
|---|---|---|
| Network isolation | Model, index and tool server in a dedicated zone with no outbound internet route | Hosting: On-prem server |
| Data minimisation | Only approved views and document sets are exposed; excluded data is never indexed | Data that must never leave the site |
| Least privilege | Read-only service identity; per-user filtering of tool calls | 250 named users |
| Model governance | Open-weight model pinned by version, weights checksummed, updates via change control | GPU: 1-2 x 48 GB class GPUs |
| Audit and retention | Append-only log of questions, tools, sources and users with an agreed retention period | Internal audit and compliance |
| Write-back control | No writes in phases 1 and 2; later writes only through supported APIs with named approvers | Risk appetite of Lakeshore Industrial Corp |
Rollout plan
| Phase | Scope | Exit criteria |
|---|---|---|
| 1. Read-only questions | Approved views and documents for a pilot group; answers show their sources | Accuracy sampled by owners; no permission leaks in review; log reviewed weekly |
| 2. Drafts | Assistant drafts emails, summaries and suggested changes for a person to copy | Users edit less over time; no unreviewed output leaves the system |
| 3. Approved actions | Selected actions executed through supported APIs after human approval | Approver named per action; rollback tested; go-live target 2027-06-30 |
Open questions for IT
1. How fresh must answers be: real time, minutes or daily? 2. Which companies, databases or libraries are in scope across 3 site(s)? 3. Who owns document permissions and how are they mapped to MAPICS and Infor XA roles? 4. Which customisations or custom tables must be exposed? 5. What is the backup, monitoring and patching plan for the GPU host? 6. Who approves any write-back and through which supported interface? 7. Is a pilot group of 10 to 20 users available for phase 1?
Approvals
Starting-point document, not a final design. Confirm table names, interfaces and security behaviour for your exact release with your vendor or implementation partner. Nothing you enter leaves your browser.
Optional XA architecture review
We can review the plan with your IBM i and XA specialists.
Optional. The tool and every export work without this. No spam.
How to use the blueprint
Enter your XA release, IBM i level, user count and sites, then describe the questions planners, buyers and customer service ask. List the data that must not leave the site. The generator produces a plan that keeps the Power system as the system of record and moves analysis, retrieval and inference to x86 servers on your network.
What XA and IBM i specifics it assumes
XA data sits in DB2 for i files that are readable over SQL. The plan assumes journaling on key files for change capture, a replica for analytical work, and a curated glossary because field names are terse. It treats the application's RPG and CL programs as the only valid write path, and it assumes a least-authority service profile.
Reading the output and common mistakes
Use the open questions for the IBM i administrator: which files are journaled, which libraries are in scope and who owns authority reviews. Typical mistakes are querying the production partition with heavy SQL, using a broadly authorised profile and writing into files directly. Adjust names to your own libraries before circulating the plan.
Frequently Asked Questions
Does the model run on the IBM i?
No. The plan places model serving, retrieval and the tool server on x86 hardware. The IBM i remains the system of record and supplies data through JDBC reads or a replica.
What if journaling is not enabled?
Then change capture needs a scheduled extract instead, or journaling must be started on the files you need. The open questions section flags this for your administrator.
Can it work air-gapped?
Yes. Choose the air-gapped hosting option and the plan keeps the replica, index and model inside the enclave, with model files brought in on controlled media.
Is the plan a replacement for migration?
No. It lets users ask questions of XA data today. A migration decision is separate, though a governed data layer can help any later move.
If you want an expert look at this plan for your XA and IBM i setup, Netray can review it. The generator itself needs no contact.
Related Tools
MAPICS and Infor XA AI Assistant Sizing and Value Calculator
Estimate GPU count, token volume, index size, hours saved and payback for a private on-prem AI assistant on MAPICS and Infor XA.
Other ERPsInfor XA (MAPICS) Application Module Reference
A lookup of the main Infor XA (MAPICS) application areas with scope and a migration consideration for each, for inventory of what you use.
On-Prem AIOn-Prem AI Security Hardening Checklist
A practical control checklist for securing self-hosted language models, covering model provenance, network isolation, data governance, host hardening, and audit readiness.
On-Prem AIOn-Prem AI Deployment Checklist
A 30-point pre-deployment checklist covering use cases, hardware, security, model operations, and rollout for self-hosted enterprise LLMs.
More Other ERPs tools
See all- Infor VISUAL Work Order Status and Structure Reference
- Infor VISUAL Manufacturing Scheduling Checklist
- Infor XA (MAPICS) Period End Close Checklist
- Infor VISUAL Database Table Reference
- IBM i Commands for Infor XA Administrators Reference
- Acumatica Generic Inquiry Design Template
- Acumatica Manufacturing Edition Setup Checklist
- Sage X3 Import Template Worksheet
- Acumatica Implementation Budget Estimator
Keep exploring
Free tools
- Infor XA Private AI Layer Blueprint GeneratorFill in your Infor XA version, database, hosting and use cases and get a shareable architecture plan for a private AI layer.
- Infor XA (MAPICS) Application Module ReferenceA lookup of the main Infor XA (MAPICS) application areas with scope and a migration consideration for each, for inventory of what you use.
- MAPICS and Infor XA AI Assistant Sizing and Value CalculatorEstimate GPU count, token volume, index size, hours saved and payback for a private on-prem AI assistant on MAPICS and Infor XA.
ERP fixes and how-tos
- How to add a private AI layer to Infor XA on-premReference architecture to add AI to Infor XA on-prem: read-only data path, retrieval, permissions, local GPUs, write-back control and rollout.
- How to Add AI to MAPICS and Infor XA on IBM i Without Leaving Your NetworkHow to give MAPICS and Infor XA on IBM i a private AI assistant: DB2 for i access, journal-based change capture, local GPU model, authority mapping and rollout.
- Infor XA costing: cost types, roll-up and standard cost setupHow product costing works in Infor XA (MAPICS): standard versus current cost, how roll-up builds cost from BOM and routing, setup order, and common pitfalls.
Integrations
- How to add an AI chatbot to Infor XAHow to add an AI chatbot to Infor XA on IBM i: grounded answers from Db2 for i, internal and customer portals, role-based access, and a fully on-prem option.
- How to add AI agents to Infor XA on IBM iHow to add AI agents to Infor XA on IBM i: a tool layer over Db2 for i and RPG programs, approval gates, on-prem models, and realistic expedite and exception flows.