I. Project Background and Objectives

1. Industry pain points

The mould is heavy (usually 0.5-10 tons), irregular in shape, low in manual handling efficiency and high in safety hazard.

The utilization rate of traditional warehouse space is less than 30%, and the mold search depends on manual records, which is easy to make mistakes.

Die life management is extensive, lacking accurate frequency of use and maintenance reminder.

2. Programme objectives

Realize the automation of the whole process of mold access, and improve the efficiency by more than 50%.

The space utilization rate was increased to 85%+, and the accuracy of inventory management reached 99.9%.

Integrated intelligent monitoring system, extending the service life of mold by 20%+.

Second, the core system design

1. Hardware system

Heavy three-dimensional shelf

Bearing range: 1-15 tons/floor, with adjustable height (0.5-3 meters for mold height).

H-beam+non-slip grille design is adopted to prevent the mold from sliding.

Customized handling equipment

Special stacker for molds:

Load 15 tons, equipped with hydraulic adaptive fixture (compatible with L-type /U-type mold).

The positioning accuracy is 1 mm, which supports double depth access.

AGV+ lifting trolley:

Used for transferring molds from production line to warehouse, equipped with laser+visual navigation.

RFID and sensor network

The mold is embedded with high-temperature resistant RFID tags (resistant to 300℃ heat treatment environment).

The shelf is equipped with a weight sensor to monitor the in-place state of the mold in real time.

2. Software system

Mould intelligent management system (Mold WMS)

Intelligent allocation module:

Automatically optimize the storage location according to the mold size, weight and frequency of use.

Example: High-frequency molds are stored in the golden location closest to the exit.

Life prediction module:

By recording punching times and stress data (connected with PLC of punching machine), the remaining life can be predicted and maintenance warning can be triggered.

3D visual monitoring:

Real-time display of mold location, inventory status and equipment operation, supporting AR remote inspection.

Integration with production system

Docking MES/ERP system to realize:

The production plan triggers the die to be pre-heated in advance.

Automatically generate mold repair orders and synchronize spare parts inventory.

Third, the implementation steps

1. Demand analysis and scheme design (2-4 weeks)

On-site mapping of die size distribution (3D scanning is recommended to generate die library model).

Formulate ABC classification strategy: Class A (daily frequency > 5 times) gives priority to automatic access.

2. Hardware deployment (8-12 weeks)

Shelf installation: ground expansion bolt and top pull rod are used for double fixation, and the seismic grade is 8.

Equipment debugging: No-load/load test of stacker (including emergency stop, anti-collision and power-off protection test).

3. System alignment and training (2-3 weeks)

Simulation of extreme scenarios: simultaneous allocation of 30 molds, system disconnection recovery, etc.

Training content: abnormal handling (such as mold stuck reset), data analysis and kanban interpretation.

Fourth, risk control

1. Technical risks

Response: The stacker is equipped with dual servo motor redundancy system, and the current task can still be completed if a single motor fails.

2. Data security

Response: Localized deployment+blockchain deposit certificate is adopted, and key operation logs cannot be tampered with.

3. Compatibility risk

Response: provide multi-protocol converters such as OPC UA/Modbus to adapt to the data interfaces of different brands of punching machines.

Program highlights:

-Combination of rigidity and flexibility: heavy hardware+flexible algorithm, which not only meets the needs of large tonnage but also adapts to many varieties.

-Life cycle management: from warehousing, use to scrapping, data-driven decision-making.

-Extensibility: Support the future access to the digital twin system of the factory, and realize the linkage between reality and reality.

If you need to refine a module (such as RFID selection list), you can further provide technical parameters.