we are committed to delivering innovative solutions that drive growth and add value to our clients. With a team of experienced professionals and a passion for excellence.

Search Now!
Follow Us

Zelar Manufacturing

Google Cloud Partner

Turn Plant Floor Data Into Manufacturing Intelligence

 

Non-invasive, platform-agnostic data pipelines from any automation vendor to Google Cloud. Real-time OEE, predictive maintenance, and AI-driven quality — deployed in weeks, not months.

 
The Challenge

Your Machines Generate Data. It Stays Trapped on the Plant Floor.

 

Mid-market manufacturers generate massive volumes of machine-level data through PLCs, SCADA systems, and sensors. But extracting that data for analytics means either invasive PLC modifications, expensive historian platforms, or $300K+ integration projects from Global Systems Integrators.

Automation Agnostic

Works With Every Automation Vendor

Our edge gateway supports 270+ industrial protocols. Connect any PLC, DCS, SCADA, or IoT platform — without vendor lock-in.

 
01.

24x7 monitoring and triage

OPC UA, EtherNet/IP, DH+

02.

Siemens

OPC UA, PROFINET, S7

03.

ABB

OPC UA, Modbus, PROFIBUS

04.

Schneider

OPC UA, Modbus TCP, EIP

05.

Yokogawa

OPC UA, HART, FOUNDATION

06.

Honeywell

OPC UA, Modbus, Experion

Also supports Emerson, Mitsubishi, Fanuc, Beckhoff, Omron, Bosch Rexroth, GE, and 250+ additional protocols including MQTT, BACnet, OPC DA/HDA, and custom gateways.

 
Reference Architecture

Six-Layer Cloud-Native Pipeline

 

From PLC tag to executive dashboard in under 10 seconds. Every layer is serverless, auto-scaling, and secured by ISA/IEC 62443 zone controls.

01

OT Data Sources

PLCs, SCADA, HMI, sensors from any vendor

02

Edge Gateway

Manufacturing Connect edge – 270+ protocols, ISA-95 normalization.

03

MDE Pipeline

Pub/Sub + Dataflow: parse, transform, enrich, deduplicate

04

BigQuery

Three-tier lakehouse: Raw, Curated, Analytics zones

05

Vertex AI

Predictive maintenance, quality anomaly detection, OEE optimization

06

Looker

OEE Command Center, maintenance, quality, executive KPIs

Step 01
01

OT Data Sources

PLCs, SCADA, HMI, sensors from any vendor

Step 02
02

Edge Gateway

Manufacturing Connect edge — 270+ protocols, ISA-95 normalization

Step 03
03

MDE Pipeline

Pub/Sub + Dataflow: parse, transform, enrich, deduplicate

Step 04
04

BigQuery

Three-tier lakehouse: Raw, Curated, Analytics zones

Step 05
05

Vertex AI

Predictive maintenance, quality anomaly detection, OEE optimization

Step 06
06

Looker

OEE Command Center, maintenance, quality, executive KPIs

Manufacturing AI Use Cases

From Visibility to Prediction

Real-Time OEE Command Center

 

Replace manual shift reports with live OEE dashboards breaking down Availability, Performance, and Quality for every machine, line, and plant. Most manufacturers discover their actual OEE is 10-20 points below their estimate.

+15-20 pts

OEE improvement

< 30 sec

Data latency (was days)

100%

Shift report automation

4-6 wks

Time to deploy

AI-Powered Predictive Maintenance

 

Vertex AI models analyze vibration, temperature, motor current, and historical patterns to predict equipment failures before they happen. Integrated with SAP PM for automated work order creation.

25-40%

Unplanned downtime reduction

15-25%

Maintenance cost reduction

50-60%

MTTR improvement

6-8 wks

Time to deploy

AI Quality Intelligence

 

Autoencoder-based anomaly detection identifies quality deviations in real-time. SPC control charts, Cp/Cpk monitoring, first-pass yield tracking, and root cause analysis traced back to material lots and process parameters.

30-50%

Scrap/rework reduction

Real-time

SPC anomaly detection

Full

Material traceability

8-12 wks

Time to deploy

Energy and Sustainability Analytics

 

Monitor energy consumption per machine, per product, and per shift. Identify energy waste patterns, correlate consumption with production output, and track sustainability KPIs for ESG reporting.

10-20%

Energy cost reduction

Per-unit

Energy cost tracking

ESG

Sustainability reporting

Add-on

Builds on Phase 1

270+

Industrial Protocols Supported

< 10s

PLC Tag to BigQuery Latency

0

PLC Code Changes Required

40-50%

Cost Savings vs. GSIs

Phased Deployment

Value at Every Phase

Each phase delivers standalone business value. Start with OEE visibility in 4 weeks, expand to predictive AI as the data foundation matures.

 

1. OEE Visibility

4-6 weeks

  • Connect 1-3 critical machines via edge gateway
  • Provision MDE cloud pipeline to BigQuery
  • Launch OEE Command Center in Looker
  • Automated shift reports replace manual collection

Baseline OEE established. Real-time visibility live.

2. Predictive AI

6-8 weeks

  • Expand to all critical equipment (10-30 machines)
  • Deploy Vertex AI predictive maintenance models
  • Integrate with SAP PM for automated work orders
  • Equipment health scores and failure predictions

25-40% reduction in unplanned downtime.

3. Quality AI + Fusion

8-12 weeks

  • Quality anomaly detection with Vertex AI
  • Full SAP Cortex IT-OT data fusion
  • Executive KPI dashboards across all plants
  • Gemini-powered natural language queries

Complete manufacturing intelligence platform.

Defense in Depth

Industrial-Grade Security

Built to ISA/IEC 62443 standards with Purdue Model zone alignment. Your plant floor remains completely air-gapped from the cloud.

01.
02.
03.
04.
05.
06.

Non-Invasive Access

Read-only OPC UA connections. Zero PLC modifications. Existing automation programs remain untouched. Deploy on running production with zero downtime.

OT DMZ Isolation

Edge gateway deployed in the OT DMZ with dual firewall configuration. Only outbound MQTT/TLS connections. No inbound access from cloud to plant floor.

Encrypted End-to-End

TLS 1.3 for data in transit, AES-256 for data at rest. mTLS authentication between edge and cloud. X.509 certificate-based device identity.

IAM + Zero Trust

GCP IAM with least-privilege service accounts. Workload Identity Federation for the edge gateway — no stored credentials on-premise. VPC Service Controls perimeter.

Audit and Compliance

Cloud Audit Logs for complete access audit trail. CMEK (Customer-Managed Encryption Keys) for BigQuery. DLP API scanning for PII in data streams.

Purdue Model Alignment

Architecture strictly follows ISA-95 / Purdue levels 0-5. Each zone boundary has explicit security controls. Plant floor operates independently if cloud disconnects.

Why Zelar

Built Different for Mid-Market

 

Google Cloud Partner delivering non-invasive, platform-agnostic manufacturing data pipelines for mid-market discrete manufacturers

01.

40-50% Below GSI Pricing

India ODC delivery model provides enterprise-grade architecture at mid-market price points. US Solutions Architect on-site; India team builds the pipeline.

02.

Automation Agnostic

270+ industrial protocols. Whether you run Rockwell, Siemens, ABB, or a mix of vendors, the same architecture works without custom integration for each.

03.

Weeks, Not Months

Rapid deployment within weeks. Non-invasive deployment means zero downtime, zero risk to running production lines. Start generating ROI immediately.

Ready to Unlock Your Plant Floor Data?

Schedule a 30-minute discovery call. We will assess your automation environment and map out a Phase 1 deployment plan — specific to your PLCs, your plant, and your goals.