How Fynd MES helped NeoLync replace five factory systems with one platform

Replacing Five Systems With One MES System

A line at a Tirupati electronics plant used to run on ERP, MES and WMS that didn't talk to each other. Forge replaces all of it with one platform, live today.

The Challenge

Electronics manufacturing plants have historically run on a patchwork of systems - separate ERP, MES, WMS and HR tools, plus a fair amount tracked in spreadsheets - none of them sharing data cleanly with each other. A single unit's serial number, IMEI or MAC address gets tracked inconsistently as it moves from PCB to assembly to QA to the shipping carton, which makes real end-to-end traceability nearly impossible.

Quality checks like SPI, AOI and X-ray inspection happen reactively rather than as part of a closed loop, procurement runs on manual vendor onboarding and PO approvals, and nobody has real-time visibility into what's actually sitting in work-in-progress versus finished goods. Problems get caught late, after the cost of rework has already been paid.

The Solution

Forge replaces that patchwork with a single, AI-native Manufacturing OS covering four stages of the factory - Plan, Buy/Source, Assemble, Move - all running on the same data spine, so every action is traceable from raw material to a dispatched carton. It's live today at NeoLync's Tirupati facility, running Jio Air Fiber Wi-Fi device manufacturing, deployed entirely on-premise for low latency and full data control.

One OS instead of five

BOM, work orders, serial-number traceability, HR and quality management all sit on a single AI-native backbone instead of siloed ERP, MES and WMS systems that don't talk to each other.

Genealogy from component to carton

Every unit gets a unique serial number or IMEI bound to it from PCB through assembly, QA, packaging and shipment, so a unit's full history is a lookup, not an investigation.

Already running real volume

Operators and line managers already work on Forge day to day, handling a steady production run-rate and absorbing sharp peaks in transaction volume during high-volume runs.

How it works

For an operator or plant manager, a unit's path through Forge looks something like this:

1. A work order gets released with its BOM, routing and staffing already planned against real capacity, not a static schedule.

2. Materials get procured and kitted, with vendor and material traceability built into that same procure-to-pay flow.

3. As the unit moves through assembly - SMT, then final assembly, then test - operators follow plain-language, step-by-step inspection instructions on a shop-floor screen, with the line stopping automatically if a check fails.

4. Every identifier gets bound to that specific unit as it goes - serial number, IMEI, MAC address and more - building a full genealogy in real time.

5. Finished units get scanned to pack, tracked through warehousing, and shipped with an ASN and e-invoice generated automatically.

6. Plant leadership watches all of it on a live dashboard - OEE, first-pass yield, active work orders, station status - rather than waiting on an end-of-shift report.

Under the hood

Eight modules sit on one shared data backbone, connected through a central API and an AI orchestration layer:

Procure-to-Pay

Vendor management, purchase orders, goods-receipt notes and approvals.

Warehousing

FIFO handling, unit-level tracking, expiry alerts and cold-storage monitoring.

MES + Production

SMT, inspection (SPI, AOI), reflow and QA, fully integrated with routing logic.

BOM & Version Control

Multi-revision bills of materials, component substitution, and full engineering-change-order traceability.

Workforce, client & integrations

People & Labour, Client Management, and Integrations - HRMS, SAP, Infor and more - round out the operational layer.

AI Orchestration

The central intelligence layer that connects and coordinates every other module, rather than sitting bolted on the side.

Still running your plant on systems that don't talk to each other?

Let's connect