Client Overview

Cahaya Electronics Manufacturing Sdn Bhd is a leading Malaysian electronics manufacturer based in the Bayan Lepas Free Industrial Zone in Penang. Operating six surface-mount technology (SMT) production lines around the clock, Cahaya produces printed circuit board assemblies and components for global consumer electronics brands, employing over 1,800 people and exporting across the Asia-Pacific region.

The Challenge

Cahaya was losing significant production to unexpected equipment failures, and a reactive maintenance culture was straining both margins and delivery commitments:

1. Unplanned Downtime: Critical SMT line failures were causing $3.2M in annual lost production
2. Reactive Maintenance: Technicians fixed machines only after breakdowns, with no early-warning system
3. Spare Parts Inefficiency: “Just-in-case” overstocking tied up $1.5M in capital while key parts still ran out
4. Quality Drift: Degrading equipment was pushing defect rates above 6% on sensitive lines
5. Scheduling Disruption: Unpredictable outages made production planning and customer commitments unreliable

The AI Solution

We deployed a comprehensive AI-powered predictive maintenance system tailored to Cahaya’s high-mix electronics manufacturing environment, combining IoT sensor data with machine learning models that forecast equipment failures before they happen.

Implementation Process

The AI solution was rolled out in phases across the Penang plant to validate accuracy on critical assets before scaling plant-wide.

Phase	Duration	Key Activities	Scope
Pilot Program	6 weeks	Sensor retrofit, baseline data collection, model training	1 critical SMT line
Phase 1 Rollout	8 weeks	Expanded deployment, technician training, alert tuning	3 production lines
Phase 2 Rollout	6 weeks	Full deployment with CMMS integration	All 6 lines + utilities
Final Rollout	4 weeks	Plant-wide optimization and supplier inventory sync	Entire Penang facility

Client Testimonial

“Before the AI system, every sudden line stoppage felt like a crisis. Now we get early warnings days in advance — we fix machines during planned downtime instead of in the middle of a production run. Our delivery reliability is up, our maintenance budget is down, and the team finally trusts the data.”

– Tan Wei Ming, Director of Operations, Cahaya Electronics Manufacturing

Key Learnings & Best Practices