1. Project Overview

Swire Coca-Cola’s Zhengzhou smart factory is one of the largest single investments in the company’s China operations — a 194-mu (approximately 13-hectare) facility with a total investment exceeding RMB 900 million (USD 125 million). With an annual production capacity of 1 million tons, this “factory of the future” is built around digitalization, sustainability, and fully automated production workflows.

At the heart of the factory’s logistics is an automated high-bay warehouse with a throughput rate of 464 pallets per hour, handling nearly 2 million cases of beverage products daily. To ensure uninterrupted operation and worker safety, the warehouse rack structure requires continuous structural health monitoring.

Key project data:

• Customer: Swire Coca-Cola Zhengzhou Smart Factory
• Location: Zhengzhou, Henan Province, China
• Facility: 194 mu (13 hectares), RMB 900M+ investment
• Production capacity: 1 million tons/year
• Warehouse throughput: 464 pallets/hour (2M cases/day)
• Sensor: ZCT330M-LWP-ANH-YKC0 NB-IoT wireless tilt sensor
• Project year: 2025

2. The Challenge: Beverage Warehouse Density Demands Continuous Monitoring

Beverage warehouses present unique structural challenges. Products are stored in high-density configurations — pallets of bottled and canned beverages are among the heaviest consumer goods in warehousing. A single pallet of canned drinks can exceed one ton. When stacked to heights of 10–15 meters across narrow-aisle racking, the cumulative load places extreme demands on rack structural integrity.

Traditional periodic inspection cannot adequately address these risks. Rack deformation caused by uneven floor settlement, impact from forklift operations, or incremental creep under sustained load may develop over weeks or months, well below the threshold detectable by visual inspection. By the time tilt becomes visible, repair costs and safety risks have escalated significantly.

The project team required a monitoring solution that:

• Operates 24/7 without human intervention
• Integrates with the factory’s existing automation systems
• Provides early warning before tilt reaches critical levels
• Deploys without disrupting ongoing warehouse operations

3. The Solution: NB-IoT Wireless Rack Monitoring System

System Architecture

Sensing layer: The ZCT330M-LWP-ANH-YKC0 NB-IoT wireless tilt sensor is mounted at the top of each monitored rack bay. The top position is selected because it amplifies small base movements into measurable angular displacement.

Communication: Data is transmitted via NB-IoT cellular network directly to the cloud platform. No local gateways, wiring, or network infrastructure is required at the warehouse. The reporting interval is adjustable from 60 seconds to 36 hours, configurable remotely based on operational risk profiles.

Alerting: When tilt exceeds the configurable threshold (default 0.5°), the system triggers instant alerts via PC dashboard and mobile app. The platform maps alerts to a floor plan overlay, enabling operators to pinpoint the affected rack location immediately.

Data logging: Historical data is stored with 5–10 minute granularity, providing full traceability for trend analysis, root cause investigation, and compliance documentation.

Key Technical Features

• Adjustable reporting interval: 60 seconds to 36 hours
• Default alert threshold: 0.5° (remotely configurable)
• Visualization: Warehouse floor plan overlay with real-time rack status
• Access: PC dashboard + mobile app
• Deployment: No wiring, no gateway, bolt-on installation
• Communications: NB-IoT cellular (SIM-based, direct to cloud)

4. Technical Specifications

5. Results & Impact

24/7 monitoring coverage: The system provides continuous structural monitoring that manual inspection cannot match. Adjustable reporting intervals allow the operations team to increase monitoring frequency during peak throughput periods.

Early warning capability: The 0.5° alert threshold catches developing rack tilt at an early stage, before deformation progresses to the point of causing product damage or creating safety hazards. The floor plan overlay enables rapid identification and response.

Zero operational disruption: Installation was completed without rack unloading, area shutdown, or disruption to the 464 pallet/hour warehouse operation. All sensors communicate wirelessly, eliminating cable routing across active warehouse aisles.

Multi-site scalability: The same system architecture has been deployed at other major manufacturing facilities including FANUC, Yili, Mengniu, and Schneider Electric, demonstrating broad applicability across industries.

6. Key Takeaways for Engineers

1. Reporting interval flexibility is critical: The ability to adjust from 60 seconds (high-risk monitoring) to 36 hours (long-term trend tracking) from a single sensor platform allows one hardware deployment to serve multiple use cases. This flexibility matters more than raw battery life numbers.
2. Floor plan integration improves response time: A sensor reading of 0.6° is just a number — until it’s plotted on a warehouse map showing exactly which aisle and bay level is affected. Visual context turns raw data into actionable operational intelligence.
3. Beverage warehouse loads are unique: The density of beverage products (1+ ton per pallet) creates rack loads that differ significantly from general merchandise warehousing. Threshold calibration should account for stored product weight distribution.
4. Multi-site architecture is straightforward: Since each sensor connects independently via cellular, adding a second or third warehouse to the monitoring platform requires no additional infrastructure. The cloud platform handles geographic aggregation natively.

7. Frequently Asked Questions

1. How does the system handle the high-moisture environment of a beverage warehouse?

The ZCT330M-LWP-ANH-YKC0 sensor is housed in an enclosure rated for industrial environments. The NB-IoT communication operates reliably in high-humidity conditions. No condensation or moisture ingress issues have been reported across the deployed installations.

2. Can the 60-second reporting interval drain the battery quickly?

The sensor’s power management system dynamically adjusts based on the configured interval. At 60-second reporting, battery life is reduced proportionally. The 36-hour long-interval mode is recommended for baseline trend monitoring, while 60-second mode can be activated temporarily during peak risk periods (e.g., floor maintenance, structural modifications).

3. Does the system integrate with existing WMS or SCADA platforms?

The cloud platform provides standard API access for integration with warehouse management systems (WMS) and supervisory control and data acquisition (SCADA) platforms. Custom integration can be configured based on the facility’s existing automation architecture.

4. What happens during a cellular network outage in the warehouse?

The sensor buffers measurement data in onboard storage. When cellular connectivity is restored, all buffered data is automatically uploaded in chronological order. This ensures no data gaps even during temporary network interruptions.