Smart Factory Planning, Integration, and Operations
This domain tests how to scope a smart-factory program, integrate MES/ERP/OT without point-to-point sprawl, and prove value in operations after go-live.
Core rules
- Start with a baseline KPI and a bounded use case. Brownfield projects must account for legacy assets, existing network segmentation, and change management; greenfield projects can standardize data models earlier but still need operations input.
- Integration should follow ISA-95 boundaries. MES/MOM owns work orders, dispatching, traceability, and quality execution; ERP owns enterprise planning and finance. B2MML/ISA-95 XML is a common ERP-to-MES message standard; OPC UA commonly connects shop-floor systems to MES.
- Prefer managed integration over point-to-point spaghetti: a broker, API gateway, or unified namespace reduces coupling. Design idempotency, retries, and reconciliation for every message flow.
- A pilot is not a rollout. Define exit criteria, scale-up funding, support ownership, and OT asset lifecycle before declaring success.
Worked example
A project costs 300,000 to implement and saves 120,000 per year, but adds 20,000 per year in support. Net annual benefit is 100,000, so simple payback is 300,000/100,000 = 3.0 years. If the benefit is only claimed, not measured against a baseline, the payback is speculative. The reasoning: include recurring costs and require a measured baseline before approving scale-up.
Traps
Piloting forever without scale criteria; skipping MES and connecting ERP directly to PLCs; using point-to-point integrations that become unmaintainable; and claiming benefits without a baseline or support model.
Key points
- Anchor projects to a baseline KPI and bounded use case.
- Respect ISA-95 ownership between ERP, MES, and control.
- Use managed integration with idempotency and reconciliation.
- B2MML is common for ERP-to-MES exchange.
- Payback must include recurring costs and measured benefits.
- Define pilot exit and scale-up criteria early.