.avif)
Shop Floor OEE Monitoring System | Sicagen
20%
40%
25%
30%
About
Sicagen India Limited is an ISO 9001:2015 certified enterprise and a trusted integrated solutions provider in infrastructure, industrial packaging, and specialty chemicals. Their drum manufacturing division produces steel drums for oil storage across a seven-stage production line — from raw sheet metal to finished drums — serving large petrochemical and industrial clients. As a high-volume manufacturer operating under tight delivery schedules, consistent machine uptime and line efficiency are critical to meeting customer commitments.
Business Challenges
Sicagen's drum manufacturing line runs across seven sequential production stages. Because each stage feeds the next, a slowdown anywhere creates a cascading backlog — but supervisors had no way to see this happening until end-of-shift counts fell short of plan.
- No real-time machine output visibility: Production counts were tracked manually at end of shift, making mid-day corrections impossible.
- Unclassified downtime: When a machine stopped, there was no way to distinguish a mechanical breakdown from a planned maintenance stop or a stall caused by waiting for output from the upstream stage.
- Plan vs. actual gap unknown intra-shift: The plant head had no live view of whether each machine was on track against its daily capacity target — problems were discovered too late to recover.
- No OEE baseline: Without continuous output and availability data, there was no reliable way to measure Overall Equipment Effectiveness, prioritise maintenance, or identify the weakest link in the seven-stage line.
- Dependency bottlenecks invisible: In a sequential line, a machine idle due to upstream material shortage looks identical to a breakdown unless sensor data distinguishes them.
Solution
We designed and deployed a hardware-software OEE monitoring system across all seven stages of Sicagen's drum manufacturing line.
Hardware layer: Installed IoT sensors on each machine to detect running state (active/idle) and count actual output units per cycle. Sensors interfaced directly with existing machine controls — no changes to production equipment required.
Software layer: Real-time data ingestion from all seven stages into a central processing engine. Automatic downtime classification — the system distinguishes between planned downtime (scheduled maintenance), unplanned breakdown (mechanical failure), and material dependency stall (idle because the upstream stage has not yet delivered input). Per-machine OEE calculation (Availability × Performance × Quality) updated live throughout each shift. Plan vs. actual output tracking per machine, with deviation alerts when a machine falls behind threshold.
Dashboard and visibility: Plant head dashboard showing all seven stages simultaneously — output count, running status, downtime type, and plan vs. actual — updated in real time. Shift-level and day-level reports for production planning and maintenance scheduling. Breakdown history and frequency analysis per machine to support preventive maintenance prioritisation.
Our Approach
Our Steps
Outcome
Deploying IoT-based OEE monitoring across the seven-stage drum manufacturing line gave Sicagen's plant leadership the visibility they needed to act on problems within the shift rather than after it.
- 40% improvement in OEE — continuous measurement and real-time feedback enabled faster recovery from both breakdowns and dependency stalls.
- 20% reduction in unplanned downtime — breakdown classification allowed maintenance teams to distinguish mechanical failures from upstream dependency issues, eliminating misdiagnosed interventions.
- 25% improvement in first-pass yield — real-time output monitoring per stage helped supervisors catch throughput drops earlier in the shift.
- 30% reduction in cycle time variance — plan vs. actual dashboards allowed line managers to rebalance workload across stages and address bottlenecks before they compounded.
- Plant head gained a single real-time view of the entire line — seven machines, live output, live status, live plan vs. actual — for the first time.
Selected Work
