# On-Prem CMMS Private AI Layer Blueprint Generator

> Source: https://netray.co/tools/cmms-on-prem-private-ai-layer-blueprint-generator
> Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems.
> Section: Free Tools

## Short answer

Fill in your On-Prem CMMS environment and get a shareable architecture plan for a private, read-only AI layer: components, controls, rollout and open questions.

## Category

maintenance

## Erp

On-prem CMMS

## Industry

Manufacturing

## Audience

- CIO
- IT manager
- Controls engineer
- Solution architect

## Tool

**Kind:** template

**Output:** document

**Fields:**

#### Company name

**Key:** company

**Type:** text

**Default:** Acme Precision Ltd

#### System version

**Key:** version

**Type:** select

**Options:**

- Maintenance Connection (on-premises)
- MP2
- Other on-prem CMMS on SQL Server
- Self-hosted open-source CMMS
- Older or unsupported version

**Default:** Maintenance Connection (on-premises)

#### Database or data store

**Key:** database

**Type:** select

**Options:**

- SQL Server
- PostgreSQL or MySQL
- Other or vendor API only

**Default:** SQL Server

#### Hosting

**Key:** hosting

**Type:** select

**Options:**

- On-prem server
- Private cloud
- Air-gapped

**Default:** On-prem server

#### Named users

**Key:** users

**Type:** number

**Default:** 80

#### Sites or plants

**Key:** sites

**Type:** number

**Default:** 1

#### Main use cases

**Key:** use_cases

**Type:** textarea

**Default:** Find similar past failures on an asset; PM compliance by craft; backlog by planner; draft a corrective work order with parts; summarise repeat failures.

#### Data that must never leave the site

**Key:** never_leave

**Type:** textarea

**Default:** Cost and labour rates, technician names, restricted site or customer assets, OEM documents under licence terms.

#### GPU available

**Key:** gpu

**Type:** select

**Options:**

- None yet
- 1 x 24 GB class
- 1 x 48 to 80 GB class
- 2 to 4 GPUs
- More than 4 GPUs

**Default:** None yet

#### Target go-live

**Key:** golive

**Type:** date

**Default:** 2027-03-01

**Blocks:**

#### Private AI Layer Architecture Plan: On-Prem CMMS

**Body:** Prepared for: {{company}}
System: On-prem CMMS (Maintenance Connection, MP2, Fiix alternatives), {{version}}
Data store: {{database}}
Hosting: {{hosting}}
Named users: {{users}} across {{sites}} site(s)
Target go-live: {{golive}}

#### 1. Context and scope

**Body:** {{company}} wants an AI assistant that works on data held in On-prem CMMS (Maintenance Connection, MP2, Fiix alternatives) without moving that data off its own infrastructure. The assistant is read-only by default. Any change to a system of record is proposed by the assistant and executed only after a named person approves it, through the system's supported interface.

Main use cases:
{{use_cases}}

#### 2. Components

**Table:**

**Columns:**

- Component
- Design for On-Prem CMMS
- Notes

**Rows:**

**0:** Data access path

**1:** Vendor-supported API or reporting views, or a reporting copy of the database; read-only account

**2:** Read-only identity, reporting copy where possible

**0:** Retrieval index

**1:** Retrieval index over work order descriptions, close-out notes, failure codes, PM procedures and OEM manuals, with asset class and site as metadata

**2:** Local vector store, metadata filters mirror permissions

**0:** Model serving

**1:** Open-weight model served with vLLM or Ollama on {{gpu}} in {{hosting}}

**2:** Size from users and questions per day

**0:** Tool server

**1:** Tool server with asset_history, find_similar_failures, pm_compliance, parts_for_asset and backlog_by_craft, each citing record numbers

**2:** Approved read actions only, parameterised

**0:** Identity and permissions

**1:** CMMS site, department and craft security carried into index metadata and enforced in the tool server

**2:** Checked before every query and retrieval

**0:** Audit

**1:** Append-only log of user, question, tool calls, query text, rows returned and answer

**2:** Retained per your records policy

#### 3. Security controls mapped to your constraints

**Body:** Data that must never leave the site:
{{never_leave}}

Hosting mode: {{hosting}}. All components run inside your network boundary. No model, index or log leaves it. For air-gapped hosting, model files, container images and updates enter through your controlled media process, and outbound telemetry in the serving stack is disabled.

Least privilege: a dedicated read-only service identity, no shared accounts, and per-user scope enforced in the tool server. Every answer shows its source records or query so a user can verify it.

#### 4. Three-phase rollout

**Table:**

**Columns:**

- Phase
- Scope
- Exit criteria

**Rows:**

**0:** Phase 1: Read-only questions

**1:** Cited answers on one site or line for a pilot group

**2:** Accuracy on a known question set accepted by the system owner

**0:** Phase 2: Drafts

**1:** Summaries, reports and proposed records for human editing

**2:** Edit rate and error types reviewed, no policy breaches in audit

**0:** Phase 3: Approved actions

**1:** Writes through the supported API or interface after named approval

**2:** Approval and rollback procedure tested, audit trail complete

#### 5. Open questions for IT and the system owner

**Body:** 1. Which interface does the vendor support for read access in {{version}}, and are there contractual limits on direct database reads?
2. Can we take a reporting copy of {{database}}, and who owns refresh and retention?
3. Which user groups map to which scopes, and who maintains the mapping?
4. What is the approved model list and the process for bringing in model updates?
5. What logging retention and review cadence does our policy require?
6. Who is the accountable owner for answer quality after go-live?

**Signatures:**

- Prepared by
- System owner
- IT security

**Notes:** This is a planning starting point, not a design guarantee. Confirm interface names, versions and vendor support terms with your vendor before building. Metrics such as PM compliance, MTBF and MTTR are computed from clean fields in SQL, with definitions aligned to SMRP-style metric definitions and an ISO 14224 style failure taxonomy where you have one. Retrieval handles the free text: similar failures, fixes that worked and procedures. New work orders are drafted by the assistant and entered by a planner.

## Content

**Intro:** This generator turns a few facts about your On-Prem CMMS environment into an architecture plan for a private AI layer that stays on your servers. It produces a document you can hand to IT, security and the system owner.

**Sections:**

#### How to use the generator

**Body:** Fill in company name, version, data store and hosting mode, then describe the use cases and the data that must never leave the site. Choose the GPU line that matches what you have or expect to buy. The document updates as you type and exports to PDF or Word. Share the first draft with the system owner and IT security, and use the open questions section as the agenda for your next meeting. Expect to edit the components table once you learn what your vendor supports.

**Bullets:**

- Start with one site or line
- Name an owner for answer quality
- List the data that must stay on site before you pick models
- Use the exported file as the basis for a change request

#### Method behind the plan

**Body:** Metrics such as PM compliance, MTBF and MTTR are computed from clean fields in SQL, with definitions aligned to SMRP-style metric definitions and an ISO 14224 style failure taxonomy where you have one. Retrieval handles the free text: similar failures, fixes that worked and procedures. New work orders are drafted by the assistant and entered by a planner.

#### What the plan covers

**Body:** The plan has five parts: context and scope, a component table, security controls tied to your inputs, a three-phase rollout and open questions. The rollout moves from read-only questions to drafted outputs to approved actions, with exit criteria between phases. Each component is described for On-prem CMMS (Maintenance Connection, MP2, Fiix alternatives) specifically, including the data access path and the permission model.

**Bullets:**

- Data access path and identity
- Retrieval index and tool server
- Model serving on local GPUs
- Audit log and retention

#### Common mistakes

**Body:** Most failed pilots skip the permission model or the audit trail, then cannot get approval to widen use. Check the points below before you take the plan to a steering group.

**Bullets:**

- Reading production tables directly without vendor confirmation, then breaking on upgrade.
- Computing MTBF or PM compliance with the model instead of SQL on clean fields.
- Letting the assistant close work orders or change PM intervals without review.

**Cta:** If you want a second pair of eyes on this plan, Netray designs and builds private, on-prem AI layers over systems like this, read-only by default.

**Faq:**

#### Does this plan require a cloud service?

**Answer:** No. Every component in the plan runs inside your network: the data access path, the retrieval index, the model serving on your GPUs, the tool server and the audit log. Air-gapped hosting is a supported option.

#### Is the assistant able to change data?

**Answer:** Not by default. The plan is read-only. Writes are proposed by the assistant and executed only after a named person approves, through the system's supported interface, and the audit log records the proposal and the approval.

#### How do I pick the GPU line?

**Answer:** Pick what you have or plan to buy, then check it with the sizing calculator for this system. A 24 GB class GPU suits small models and few users. More users or larger models need more memory or more GPUs.

#### Can I edit the document?

**Answer:** Yes. Edit the fields and export to Word. The component table and the open questions are the parts most worth tailoring once your vendor confirms which interfaces it supports for your version.

## Search terms

on-prem cmms ai architecture, on-prem cmms on-prem ai, private ai layer blueprint, on-prem ai architecture template, air-gapped ai architecture plan, local llm architecture document, read-only ai assistant design

---
Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json