How RF-SMART Process Manufacturing connects to Oracle JD Edwards EnterpriseOne
Native E1 Process Manufacturing transactions
Per the vendor, handheld and shop floor transactions post through EnterpriseOne's own Process Manufacturing logic rather than direct table writes. That keeps costing, inventory and work order status consistent with how JDE expects them. Confirm which JDE releases and process manufacturing programs are supported.
What data moves
| From | To | Data | When |
|---|---|---|---|
| JD Edwards | RF-SMART | Process work orders, ingredients, co-products and by-products | Work order release or lookup on the device |
| RF-SMART | JD Edwards | Co-product and by-product completion quantities with lot | Operator confirms completion |
| RF-SMART | JD Edwards | Scrap quantity and reason | Scrap entry on the device |
| RF-SMART | JD Edwards | Hours and labor against the work order or step | Start/stop or time entry |
Setting it up
- 1Confirm scopeList the process work order types, plants and steps where operators will report, and which transactions are needed.
- 2Prepare JDE process dataCheck recipes, co-product and by-product definitions, units of measure and lot control before devices go live.
- 3Set up users and securityCreate device users mapped to JDE users and limit them to the transactions they need.
- 4Configure scrap reasons and labor codesAlign scrap reason codes and labor types with costing rules.
- 5Deploy devices and labelsProvision handhelds, test Wi-Fi coverage and print lot or container labels where used.
- 6Pilot a lineRun real batches, compare quantities and cost in JDE with paper records, then roll out.
What usually breaks
- !Co-product yield percentages in JDE that do not match reality cause cost distortion when operators report actuals.
- !Unit of measure differences between recipe and inventory units create rounding errors at completion.
- !Lot attributes such as expiry or potency missing at completion break later traceability.
- !Weak Wi-Fi in tanks or cold rooms leads to unsynchronized transactions and duplicate entries.
- !Unreleased or closed work orders reject transactions with unclear messages on the device.
A private AI layer across both systems
Yield variance review
A private assistant can compare reported completions and scrap with standard yields in JDE and list batches with unusual loss, with the data shown.
Rejected transaction triage
Explain in plain language why a device transaction failed (closed order, wrong lot status) and draft the correction for a supervisor to approve.
Genealogy questions
Answer which finished lots used a given ingredient lot, from JDE lot records, for quality investigations.
Labor anomaly flags
Flag work orders with hours far outside history for review.
Try it here, free
JD Edwards EnterpriseOne AI Assistant Sizing and Value Calculator
Estimate peak load, GPUs, index size, hours saved and payback for an on-prem AI assistant on JD Edwards EnterpriseOne.
When something else fits better
- JDE's own Process Manufacturing screens or Work Order Time Entry are enough when volumes are low and a terminal sits at the line.
- A dedicated MES may be a better fit when you need recipe execution, weighing and electronic batch records.
- Paper capture with back-office entry works for low-volume operations but delays costing.
Frequently asked questions
Does RF-SMART write directly to JDE tables?
The vendor describes native E1 Process Manufacturing transactions. That implies posting through JDE business logic rather than raw table inserts. Ask for the exact interface for your release so you know how upgrades are handled. Test unusual cases such as scrap, rework and negative yields in a sandbox before operators rely on the device in production.
Can it handle co-products and by-products?
Yes, recording co-product and by-product completions is a stated feature, along with scrap and hours and labor. Your JDE recipes must define those outputs correctly first. Check how by-product costing is configured in JDE, because device quantities only help if the cost distribution rules are right.
Can we add AI on the plant network only?
Yes. A model hosted on your own servers can read JDE and device logs to explain failures and flag yield issues, with no data leaving the site. Any correction posted to JDE needs human approval. This is useful for security-sensitive plants, since recipes and yields are often treated as trade secrets that should stay in house.
Pairing documented at www.rfsmart.com/jd-edwards-enterpriseone/solutions/process-manufacturing . RF-SMART Process Manufacturing is a product of RF-SMART; Netray is independent and not affiliated with either vendor.