Migrating from Kepware to MDCplus: Step-by-Step Guide for Integrators

Switching connectivity software mid-project is a real undertaking, but on CNC monitoring deployments the math usually works out - see Why Integrators Switch from Kepware. This guide walks through the proven migration path, designed to keep customer data flowing and dashboards live while you transition.

Quick Answer: Typical Migration Timeline

These ranges assume CNC-only migration. PLC and SCADA portions stay on Kepware unless explicitly in scope. Multi-site adds 1–2 weeks per additional site for coordination.

Pre-Migration: Inventory and Audit

Tag Inventory Export from Kepware

KEPServerEX exports the entire project as XML or CSV via the configuration tool. For each machine, capture:

• Connection details (IP, port, controller series).
• Tag list with names, data types, scan rates.
• Any calculated tags or aliases.
• Alarm and event configurations.
• IoT Gateway endpoint configurations.

Save the export as the migration baseline.

Identify Custom Logic and Calculated Tags

Plain Focas/OPC UA tags map mechanically. Anything customer-built - calculated tags using KEP arithmetic, custom alarms, derived OEE estimates - needs explicit re-implementation in MDCplus. List these now; re-implementing them is the bulk of the engineering work in migration.

Map Customer's Downstream Consumers

Who consumes Kepware data today? Historian (which one), MES (which connector path), BI tool (which dashboard), MQTT broker, custom apps. Each consumer needs a switching plan - either point at MDCplus's API instead, or have MDCplus mirror the existing OPC UA / MQTT endpoint structure.

This step is most often underestimated. A consumer you forgot about will surface as an angry user in week 6.

Phase 1: Parallel Deployment

Install MDCplus Edge Alongside Kepware

MDCplus deploys to the same network segment as Kepware. Same machines reachable, no firewall changes for outbound paths. For air-gapped customers, run on the same OT-side VLAN as KEPServerEX.

Both systems can poll the same controllers simultaneously without interference for most CNC protocols. Verify per-protocol - some Heidenhain DNC sessions don't support multiple concurrent clients.

Connect a Pilot Group of Machines

Start with 3–5 machines representing the brand mix. Configure them in MDCplus, validate data flow, compare against Kepware values for the same data points.

Validate Data Parity

Spindle load, axis positions, modal codes, alarms - verify MDCplus values match Kepware values for the pilot machines. Differences usually trace to:

• Different sample rates.
• Different scaling/units.
• Calculated vs raw tags.
• Time synchronization.

Fix in MDCplus configuration. Don't proceed until parity confirmed.

Phase 2: Tag Translation

From Kepware Tag Tree to MDCplus Model

MDCplus's data model is pre-defined per CNC brand. Kepware tag paths like Channel1.Device1.Spindle.Load map to MDCplus's Machine[X].Spindle.LoadPercent. Most maps are mechanical.

For consumers expecting specific tag paths (existing dashboards, MES connectors), MDCplus can expose data via OPC UA with custom tag tree matching the legacy path. This avoids forcing downstream systems to re-map.

Handling Calculated Tags and Aliases

Calculated tags (e.g., RunningRatio = (Cycle / TotalTime) * 100) translate into MDCplus's calculated metrics layer. Functionally equivalent, syntax different. Each calculated tag is a small engineering task - not heroic, but real.

Preserving Historian Continuity

If the customer has a historian continuously consuming from Kepware, plan the cutover so historian doesn't see a gap or a duplicate window:

Option A: switch historian feed at exact cutover moment, accept few-second gap.

Option B: backfill MDCplus data into historian for the parallel-run period.

Option C: MDCplus and Kepware both feed historian during parallel run; deduplicate post-cutover.

Discuss with customer's data team early.

Phase 3: Full Cutover

Cut Over by Cell or by Brand

Don't attempt a "big bang" cutover for any fleet larger than 15 machines. Phase by:

• Cell (one production cell at a time).
• Brand (Fanuc first, then Siemens, then Mazak, then Heidenhain).
• Customer-defined batches.

Each phase = pilot validate → switch consumers → decommission Kepware for that batch.

Downstream System Reconfiguration

Each consumer gets reconfigured to point at MDCplus:

• Historian: point ingestion at MDCplus OPC UA or MQTT.
• MES: update connector source. - BI dashboards: update data source URL.
• Custom apps: update API endpoint.

Test each downstream after each batch.

Customer Validation

Customer signs off per batch. They validate:

• Data is flowing.
• Values match prior baseline.
• Dashboards work.
• Their team can use the new UI.

Don't move to next batch until validated.

Phase 4: Decommission Kepware

After all batches cut over and validated for 1–2 weeks:

• Stop KEPServerEX service.
• Wait one week monitoring downstream alerts.
• Uninstall and archive configuration.
• Document what was removed.

Keep Kepware license active for one renewal cycle in case rollback needed (rare but useful insurance).

Common Pitfalls and How to Avoid Them

• Forgotten downstream consumer. Audit thoroughly before cutover.
• Calculated tags assumed to "just work." They never do; budget engineering time.
• Time synchronization differences. Verify NTP across MDCplus and Kepware before parity comparison.
• Customer hasn't decided who owns the new dashboards. Define ownership early.
• Mid-renewal cycle on Kepware. Plan migration so you don't pay another year of PSP unnecessarily.
• Heidenhain DNC single-client limitations. Schedule the parallel run carefully or sequence.

Migration Project Cost and Timeline

For a representative 30-machine migration:

• Pre-migration audit: 20 hours.
• Parallel deployment: 40 hours.
• Tag translation and validation: 80 hours.
• Phased cutover (3 batches): 60 hours.
• Customer training and handover: 20 hours.

Total integrator labor: ~220 hours.

Software: MDCplus per-machine licenses for the migrated fleet.

Customer benefit: reduced ongoing software cost, working analytics, dashboards in their hands within weeks. Typical financial payback inside 12–18 months.

Frequently Asked Questions

Will the customer see downtime during migration?

No, if done correctly. Parallel run avoids any data gap; cutover is controlled per batch. Customer's operators see no difference.

Can MDCplus expose the same OPC UA tags Kepware did?

Yes. MDCplus can publish OPC UA with custom tag tree matching legacy Kepware paths, so downstream consumers don't need reconfiguration.

What if the customer's MES expects specific Kepware-format data?

MDCplus's API can mirror Kepware's data shape for MES consumption, avoiding MES-side changes.

Do I keep Kepware running indefinitely?

Recommended: keep license active for one full renewal cycle (12 months) post-migration as rollback insurance. Then decommission.

What about Kepware drivers I built custom logic against?

Custom KEP code re-implements as MDCplus calculated metrics or scripts. Functionally equivalent, syntax different. List these tags during pre-migration audit.

How do I handle a customer who's nervous about the switch?

Pilot with one cell. Show side-by-side data parity. Let them see the new dashboards while old ones still work. Confidence builds during parallel run.

Can MDCplus engineering help with migration?

Yes - MDCplus offers migration assistance to integrator partners including tag-mapping templates from Kepware exports.

When should I NOT migrate?

If the project is heavy PLC/SCADA scope, Kepware fits better - stay. If customer is committed to PTC ecosystem (ThingWorx, Vuforia), Kepware fits better - stay. Migration only makes sense when CNC-specific patterns from Why Integrators Switch apply.

Talk to MDCplus Engineering About Your Migration

For your specific Kepware deployment, request a demo - we'll review your tag inventory, scope migration effort, and provide a written migration plan within 5 business days. Or try MDCplus free on one machine first to validate fit.

Related:

• CNC Connectivity Software for Integrators (Hub)
• Why Integrators Switch from Kepware
• MDCplus vs Kepware
• Kepware Alternatives
• OEM Protocol Licensing
• Project Cost Calculator
• Buying Checklist
• Why MDCplus
• Focas Tech Guide for Fanuc

About MDCplus

Our key features are real-time machine monitoring for swift issue resolution, power consumption tracking to promote sustainability, computerized maintenance management to reduce downtime, and vibration diagnostics for predictive maintenance. MDCplus's solutions are tailored for diverse industries, including aerospace, automotive, precision machining, and heavy industry. By delivering actionable insights and fostering seamless integration, we empower manufacturers to boost Overall Equipment Effectiveness (OEE), reduce operational costs, and achieve sustainable growth along with future planning.