Manufacturing SystemsERP Platform

Opcenter APS + Preactor + private AI

AI for Siemens Opcenter APS and Preactor Installations

Short answer

AI for Siemens Opcenter APS, including long-running Preactor installations, adds a private assistant that explains schedule outcomes, traces late orders to their constraint, and reconciles the schedule with ERP and MES data. It reads through governed connectors and leaves the scheduling engine and any release to the floor under planner control.

ERP: Siemens Opcenter APS, Preactor
Industries: Manufacturing, Aerospace, Electronics, Automotive
Written for: Production Planner

Discuss your Siemens Opcenter APS environment See use cases

Many plants run Preactor today under its newer Siemens name, Opcenter APS, and some run versions that predate the rebrand. The pattern is similar either way: the ERP supplies orders, routings and stock; the APS applies finite capacity, secondary constraints such as tooling or operators, and priority rules to build a sequence; the result goes back to the ERP or to an MES or the floor as dispatch lists.

Installations tend to be deeply customized. Resources, calendars, constraint definitions, sequencing rules and data import mappings were set up by an integrator or a long-serving planner. Those choices decide what the Gantt shows, and they are often the part nobody has written down. When someone asks why an order sits behind another, the answer lives in rule settings and import mappings.

Opcenter APS and the wider Siemens portfolio provide their own analytics and visualization, and for many sites that is sufficient. A private AI layer helps when planners need explanations and cross-system answers, when the APS must be compared with live ERP or MES data, and when order and customer data cannot be sent to a public service.

This page lays out how that AI layer connects to an Opcenter APS or Preactor estate, which questions it handles well, and where it adds nothing beyond what the product already does. It does not alter the scheduling engine.

What usually gets in the way

The problems we hear most from production planner teams running Siemens Opcenter APS.

Legacy mappings nobody documented

Import and export mappings between the ERP and the APS were built years ago. When an order shows wrong, tracing it through the mapping takes a specialist.

Secondary constraints hide the real bottleneck

Tooling, operators, materials and setup groups can be the binding constraint rather than the machine. Spotting which one drove a delay takes digging through the schedule.

Upgrade and rebrand uncertainty

Sites on older Preactor releases face decisions on moving to current Opcenter APS. Knowing exactly what is customized is a prerequisite and is rarely documented.

Schedule versus execution gap

The plan assumes durations and availability; the floor reports something else through the MES or ERP. The gap is visible but explaining it is manual.

Data sensitivity

Orders, programs and customer delivery dates are commercially and contractually sensitive, which limits use of public AI tools.

Where AI earns its place in Siemens Opcenter APS

Each use case names the ERP objects it reads or writes, so your ERP team can judge the integration effort before anyone commits budget.

Why is this order late?

Trace a late order through its operations to the constraint that delayed it, whether machine, tooling, operator or material, and describe it in plain language.

Touches: Orders, Operations, Resources, secondary constraints, calendars

Outcome: Gives planners and sales a consistent explanation and highlights constraints worth relieving.

Schedule versus actual

Compare planned operation start and end times with confirmations from the ERP or MES and summarize where the plan is consistently optimistic.

Touches: Scheduled operations, ERP confirmations, MES events

Outcome: Supports better duration and efficiency assumptions in the model, with evidence rather than impressions.

Import and mapping diagnosis

When an ERP order appears incorrectly in the APS, check source data, mapping rules and import logs to suggest where it broke.

Touches: Import mappings, staging tables, ERP production orders, logs

Outcome: Shortens troubleshooting that otherwise waits for the integrator.

Rule and configuration documentation

Read resource definitions, sequencing rules and constraint settings and produce a searchable description of how the plant is modeled.

Touches: Resource definitions, rules, calendars, constraint tables

Outcome: Reduces key-person risk and prepares the ground for an upgrade or migration.

What-if summary

Summarize the effect of a changed calendar, extra shift or resource outage on late orders and load.

Touches: Calendars, Resources, Orders, scenario copies

Outcome: Speeds up decisions on overtime and subcontracting with visible evidence.

Material shortage view

Combine the schedule with ERP inventory and purchase data to list which scheduled jobs depend on parts not yet covered.

Touches: BoM, inventory, purchase orders, scheduled operations

Outcome: Lets buyers and planners work from the same shortage list.

Floor Q&A

A supervisor asks what is next on a work center and why, with answers drawn from the dispatch list and ERP status.

Touches: Dispatch lists, work center data, ERP order status

Outcome: Cuts routine interruptions to planners.

Reference architecture

The assistant reads APS data and ERP or MES data via governed, read-only connectors and produces explanations and draft proposals. The scheduling engine, rules and release process stay unchanged.

1
APS and ERP connectors
Reads APS data from its database or data exports and ERP or MES data through APIs, ODBC or SQL views, using scoped service accounts. The connector is tested against your release.
2
Semantic layer
Maps orders, operations, resources and constraint names, including site-specific fields, to terms planners use.
3
Model serving
Open-weight models served with vLLM or Ollama on customer hardware or a private cloud.
4
Retrieval and agents
Retrieval over live data and configuration documentation; proposals are drafts for planners to apply.
5
Governance and audit
Question, query and answer logging per user, with data retained inside your environment.

Integration notes for your ERP team

Preactor and Opcenter APS commonly exchange data with the ERP through ODBC or SQL staging tables, files, or an integration layer; point the AI at the same staging data rather than creating a parallel feed.
Use a read-only database role on the APS data store; never give the assistant rights to modify rules or release schedules.
Identify which release you run, since older Preactor versions and current Opcenter APS differ in data stores and interfaces.
Index import mapping definitions and logs, because import issues are the most common planner question.
Map custom constraint and resource attributes in the semantic layer or answers will miss site-specific logic.
For ERP data, use documented APIs or read replicas and respect the ERP's role model.
MES confirmations, when available, are read to compare plan with actuals; keep the MES connection read-only.

Deployment options

Air-gapped on-prem

Aerospace, defense and automotive plants with strict customer data rules.

All components run on-site next to the APS server with no outbound connections.

Private or sovereign cloud

Sites that host Siemens and ERP workloads in their own tenant.

Deployed in the same region and network boundary as the existing systems.

Hybrid

Groups mixing legacy Preactor plants and newer Opcenter sites.

One governance layer, per-site connectors, data stays at each site.

Compliance and data control

How the architecture supports your obligations. Certification and accountability stay with your organisation; the design keeps the evidence straightforward.

ITAR / EAR data handling

Run the model on infrastructure you control so controlled program data stays within your access-controlled environment and your own personnel rules.

AS9100 planning records

Reproducible, logged explanations of schedule decisions support control of production planning records.

Customer flow-down clauses

No order data reaches public AI services, which matches typical prime contractor requirements.

GDPR / regional privacy

Personal data about operators and customers remains under your control.

Where Netray fits

ERPray

Grounded answers over ERP orders, stock and purchasing, so planners can see the supply side next to the schedule.

DataRay

Useful when APS tables, exports, mapping documents and MES data sit in different sources and must be queried together on-prem.

Custom build

Constraint tracing and plan-versus-actual agents depend on your specific model and are typically built for the site.

How an engagement runs

Phase 1 . 2-4 weeks

Discovery

-Inventory of APS release, data stores and interfaces
-Mapping and rule documentation gaps
-Prioritized planner questions
-Deployment recommendation

Phase 2 . 6-8 weeks

Pilot

-Read-only connectors on a non-production copy
-Late-order tracing and plan-versus-actual live for a pilot team
-Accuracy review against planner-verified answers
-Logging validated

Phase 3 . 4-6 weeks

Production

-Production cutover with scoped accounts
-Runbook for new questions and fields
-Planner training
-Audit log review

Phase 4 . Ongoing

Scale

-More plants and data sources
-Documentation support for upgrade planning
-Model refresh
-Quarterly accuracy review

Questions to ask any vendor, including us

A short list that separates real Siemens Opcenter APS AI work from a chatbot demo.

Which APS release and data store does the connector support, and what happens at upgrade?
Can the assistant ever modify rules, calendars or released schedules?
How does it handle our custom constraints and import mappings?
Where does the model run, and what leaves our network?
How is accuracy measured against a planner's own investigation?
Can it read MES confirmations without write access?
How do you document the mapping so we are not dependent on one person?
What do we keep if the engagement ends?

Frequently asked questions

Does this work with older Preactor versions?

Often yes, provided the data store and exports can be read safely, but it depends on the release and how the integration was built. Discovery identifies the version, data store and interfaces first, and confirms what can be read before any commitment.

Does the AI change the schedule?

No. The scheduling engine and rules remain as configured, and release to the floor stays a planner action. The assistant explains results, traces causes and drafts proposals.

Why not just use Siemens' own analytics?

If built-in dashboards answer your questions, use them. The AI layer is for explanatory, cross-system questions such as why an order is late given ERP stock and MES actuals, and for sites that need the model running inside their own network.

Can it help with an upgrade from Preactor to Opcenter APS?

It can help document how the current model is built: resources, rules, constraints and mappings. That documentation reduces risk, but the migration itself remains an APS project with its own tooling and specialists.

How does it connect to the ERP?

Through documented APIs, ODBC or SQL views, or a read replica with a scoped read-only account, so ERP permissions still apply. The same staging data used by the APS integration is a natural source.

Is data safe for defense work?

The architecture keeps data and model inside infrastructure you control, so you can apply your own ITAR and customer-specific controls. Compliance obligations remain yours, and the design is meant to make them easier to meet.

How long before we see results?

A focused pilot of six to eight weeks after two to four weeks of discovery is typical. Complex legacy mappings or several plants extend that.

Related guides

Forecasting plus a reason why

AI Demand Forecasting for Manufacturers: Better Numbers and an Explanation Planners Can Trust

AI demand forecasting explains ERP forecast changes in plain language, flags exceptions for planners, and handles short-history SKUs without Excel overrides.

SAP PP/MRP + AI agents

AI Agents for SAP Production Planning, MRP, and Exception Triage

AI agents that triage SAP MD04 exceptions, explain predictive MRP shortages, and draft planned-order actions, grounded on your own SAP data, deployed on-prem.

Shop floor + on-prem AI

AI shop floor assistant for operators working inside your ERP

An on-prem AI copilot answers operator questions against your ERP work instructions, travelers, and routings in plain language, at the machine, without a screen full of menus.

Digital thread + AI

AI Across PLM and ERP: Teamcenter, Windchill, and Your ERP

AI that reads Teamcenter or Windchill alongside your ERP to trace engineering change impact, BOM sync gaps, and where-used questions, on-prem.

RAG + SQL + permissions

A 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.

AS9100D + on-prem AI

AI for AS9100 Quality Management on Your ERP

AI on top of your ERP quality module for AS9100D suppliers: NCR/CAPA drafting, FAI support, counterfeit parts screening, with a full audit trail.

Plan it with numbers

Free Tool

Talk it through with an engineer who knows Siemens Opcenter APS

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.

Start the conversation See ERPray

All AI for ERP guides