Industrial AI for New Zealand Manufacturing and Energy
Operators are now putting AI agents on top of their historian data. Instead of engineers hunting through trends after every event, the agent gathers the context and delivers the answer. I combine deep PI System experience with hands-on LLM engineering to build this for New Zealand plants and networks.
Where industrial AI pays off
- Engineers spend hours collecting data from several systems after every process event.
- Critical know-how sits with a few experienced people and in procedures nobody can find.
- Events are detected automatically, but the investigation is still manual.
- Shift handovers and production reports are written by hand every day.
- AI vendors pitch new platforms that ignore your existing PI System investment.
What I build on operational data
This is no longer experimental. Global operators presented AI agents built on PI System Asset Framework at AVEVA World 2026. The opportunity is to apply the same approach in New Zealand, on the data you already collect.
Event investigation agents
When PI AF detects an event, an agent assembles the related process data, history and procedures and drafts the report for an engineer to review.
Plain-English access to operational data
Engineers and managers ask questions in plain English and get answers grounded in PI data, with the trends to back them up.
Knowledge assistants for operations
AI that answers from operating procedures, maintenance records and engineering documents, with sources cited.
Automated shift and production reporting
Reports drafted automatically from historian and system data, reviewed by people rather than written by them.
AI-ready PI Asset Framework
Getting AF structure and data quality right, the foundation every industrial AI use case depends on.
Use cases and business cases
Prioritised industrial AI opportunities, with costs and ROI quantified in terms a board will approve.
My End-to-End Approach
One senior person accountable from the first workshop to measured results. No handoffs between analyst, architect and developer, so nothing gets lost in translation. Engage me for the whole journey, or for any single stage.
-
01
Understand
Business analysis first: stakeholder workshops, current-state process mapping and the real problem behind the request.
You getProblem statement and current-state map -
02
Quantify
Size the prize: what the current way of working costs, options compared, and ROI in terms a board can approve.
You getBusiness case and prioritised options -
03
Design
Requirements, data and solution design, choosing the right answer independently: AI, automation, reporting or simply a better process.
You getRequirements and solution design -
04
Build
Hands-on delivery of AI agents, automation, Power BI and integrations, using AI-assisted engineering to show working results early.
You getWorking solution, tested on your data -
05
Embed
Change support, training, documentation and handover, so your people actually use it and can run it.
You getA solution your team owns -
06
Measure
Track the benefits against the business case and improve whatever is not delivering.
You getMeasured results
A rare combination
Most AI consultants have never worked inside a plant, and most PI specialists have never built an LLM application. I spent years inside Methanex New Zealand, Winstone Pulp International and Powerco working with operational data and the PI System, and I build AI products myself, including Semantic Robot and AIGuardU. One person can take you from historian tag to working AI agent.
Every engagement is delivered by Luis Silva personally. No juniors, no subcontractors, and no vendor commissions.
Ways to work together
Questions
Do we need a new AI platform?
Not necessarily. Most value comes from building on your existing PI deployment and Microsoft environment. Where you already have a data platform, I integrate with it rather than replacing it.
Is it safe to let AI work with plant data?
The agents read, analyse and draft. Control actions stay with people and your control systems. Access is read-only and scoped, and engineers review what the AI produces.
What do we need in place before starting?
A reasonably structured PI Asset Framework and a clear, costly problem. If AF needs work first, I can assess readiness and fix the foundations as part of the engagement.
Can results show up in the tools our people already use?
Yes. Outputs can be delivered into Microsoft Teams, email, Power BI or PI Vision, so nobody has to learn a new system.
Tell me your biggest bottleneck
Book a 30-minute process review. No commitment and no sales pitch, just an honest view of whether and how it can be fixed.
Book a 30-Min Process Review