How to build a mobile app for JD Edwards EnterpriseOne
Short answer
Use the AIS Server and Orchestrator as the API layer for a purpose-built mobile app, or evaluate Oracle's own mobile and tablet-friendly options first. Scope the app to a few high-frequency tasks such as scanning, picking, labor and approvals, and design for scanner hardware and spotty Wi-Fi.
JD Edwards EnterpriseOne has a browser client that works on tablets, and Oracle has offered mobile and simplified UX options over the years. For many shop floors that is still not enough: operators want to scan a bin label, confirm a quantity and move on, in a warehouse where the Wi-Fi drops behind racking.
The practical question is which tasks justify a purpose-built app. The answer is usually a short list: inventory transfers and cycle counts, receiving against POs, picking and shipping confirmation, work order labor and completion, and approvals for managers who are rarely at a desk.
AIS and Orchestrator make this tractable because you can expose each task as a narrow, validated REST call. The app should not replicate the JDE UI; it should do three or four things quickly and defer everything else to the web client.
What teams use it for
Inventory transfers and cycle counts
Scan from and to locations and the item, then call a transfer or count orchestration that respects branch/plant and lot controls.
Receiving against purchase orders
Scan the PO, enter quantities per line, capture lot and expiry, and post the receipt, with over-receipt tolerance enforced.
Work order labor and completion
Operators clock on and off a work order step, report quantity complete and scrap, and attach photos.
Pick and ship confirmation
Show open pick lines for a user, confirm scanned picks, and trigger ship confirm when the order is complete.
Manager approvals
Provide a small queue of requisitions, voucher holds and change orders with approve or reject.
Maintenance rounds
Scan asset tags, view open work orders for the asset (F4801), and log readings or completion notes.
JD Edwards EnterpriseOne integration surface
- AIS Server REST: Form Service, Data Service and Orchestrator calls; the usual backbone for custom mobile clients. Exact endpoints and token handling depend on your Tools release.
- Orchestrator Studio for packaging a task (for example a warehouse transfer) as one call, with inputs validated and business logic remaining in JDE applications and business functions.
- Inventory tables for context: F41021 (item location), F4111 (item ledger), F4108 (lot master), F4101 (item master). Prefer service calls over direct table writes.
- Work order and labor processing around F3111, F3112 and time entry flows; confirm the processing options your site uses.
- Barcode and label printing from JDE reports, plus device hardware support through the scanner's keyboard wedge or SDK.
- Oracle-provided mobile or simplified web options, which vary by release and should be evaluated before building from scratch.
How it works
- 1
Pick the three tasks
Observe the floor and choose tasks done hundreds of times a day. A narrow app that works beats a broad one nobody trusts.
- 2
Choose the client technology
A progressive web app is the cheapest to deploy and works with camera scanning; a native or cross-platform app is better for rugged scanners and robust offline storage.
- 3
Design orchestrations per task
Each call accepts barcode-derived inputs and returns clear errors, such as location blocked, lot expired or quantity exceeds open. Keep business rules in JDE.
- 4
Authenticate sensibly
Use named JDE users or tokens with single sign-on where available, not a shared kiosk login. Badge scan login helps on shared devices.
- 5
Plan for offline
Queue transactions with unique IDs, sync when connected, and show clear states: pending, accepted, rejected. Block tasks where stale data would cause harm, such as picking against changed stock.
- 6
Pilot on one area
Run it in one warehouse zone or line, measure keystrokes saved and errors, then widen. Keep the web client as the fallback.
Example flows
Bin-to-bin transfer
Trigger: An operator scans a source location label.
- The app calls an orchestration to list items and quantities at that location.
- The operator scans the item and enters a quantity, then scans the target location.
- The transfer orchestration validates location type, lot and availability, and posts the move.
Result: A transfer takes seconds, and the item ledger reflects it immediately, with no paper tickets.
PO receipt with lot capture
Trigger: A receiver scans the PO barcode on the packing slip.
- The app lists open lines with expected quantities.
- The receiver enters quantity, scans or types the lot, and optionally photographs the packing slip.
- The receipt orchestration enforces tolerance and posts, or returns a clear reason if blocked.
Result: Receipts are posted at the dock rather than hours later from paperwork.
Labor reporting during an outage
Trigger: Wi-Fi drops in a machining bay.
- The app stores clock-on and completion events locally with timestamps.
- When the device reconnects, events are replayed in order.
- Rejected events appear in a correction list for the supervisor.
Result: No lost labor records and a visible trail of what needed correction.
Keeping data on-prem
Run AIS inside your network and let devices reach it over the plant Wi-Fi or a VPN, optionally through a reverse proxy in the DMZ for remote sites. Nothing needs to go to a public cloud. If you want voice or natural-language entry, such as saying a part number or asking what is at a location, a locally hosted open-weight speech and language model can run on a GPU server in your data centre. For air-gapped plants, distribute the app through your own device management system rather than a public app store.
Security checklist
- Use individual logins or badge-based authentication; avoid shared accounts so JDE audit trails stay meaningful.
- Enforce device management, screen lock and remote wipe for phones and handhelds.
- Keep AIS behind a proxy or VPN and allow only the orchestrations the app needs.
- Make every write idempotent using a client transaction ID to avoid double posting after retries.
- Do not cache sensitive data on the device longer than necessary, and encrypt local storage.
Ways to get it done
Evaluate Oracle's mobile or simplified UX offerings and the browser client
Effort: Configuration time; limited scanning and offline behaviour.
Good if your tasks match the standard applications and you lack developers.
Third-party warehouse or shop-floor mobility product with a JDE connector
Effort: Weeks to months; licence and integration cost.
Strong for scanning-heavy warehouses with standard processes.
Custom app on AIS built in-house or with Netray
Effort: A pilot of a few tasks in weeks, with ongoing support.
Best for specific workflows, on-prem privacy and later AI features like voice or photo capture.
Pitfalls
- !Trying to reproduce full JDE screens on a phone instead of three focused tasks.
- !Ignoring offline behaviour until the first dead zone causes duplicate transactions.
- !Letting the app write directly to inventory tables rather than going through JDE business logic.
- !Underestimating label quality and scanner configuration work.
- !Shared logins that make it impossible to audit who moved stock.
Frequently asked questions
Does JD Edwards have an official mobile app?
Oracle has provided mobile and tablet-friendly options for EnterpriseOne at different times, and the web client works in a browser. What is available depends on your Tools release and licensing, so check current Oracle documentation before building your own.
What API should a custom mobile app use?
The AIS Server is the usual choice, using Orchestrator endpoints for task-level calls. It keeps business logic and security inside JDE rather than in the app.
Can the app work offline?
Partly. You can queue low-risk transactions and sync later, but tasks that depend on current stock or status should require a connection. Design clear states for pending and rejected items.
Which devices work best for scanning?
Rugged Android handhelds with built-in scanners are typical on warehouse floors, while phones and tablets with camera scanning suit lighter use. Test with your actual label types and lighting.
Can we add AI features to the app?
Yes, for example photo capture of damage, voice lookups or natural-language questions about stock. Keep them read-only or review-based at first, and run models on your own hardware if data must stay on-prem.
More for JD Edwards EnterpriseOne
Want Mobile app working on your JD Edwards EnterpriseOne?
A free 30 minute assessment: we map the integration path for your version and setup, and where it should run.
Get a free assessment