The disruption of the retail industry by technology has reached the level of science fiction. You don’t have to go through the hassle of gathering all the Dragon Balls. With a phone that has downloaded the corresponding app, you can “summon” convenience stores to your side.
The reason for the popularity of unmanned retail containers is that the retail industry in China has too high costs, including high warehousing and procurement costs; The true significance of unmanned containers lies in the data they collect; At present, the convenience retail industry is a slow growing market, and in a slow growing market, what is being seized is actually the existing market share.
Architecture of Unmanned Retail System

• Machine side: The device synchronizes the software or hardware failures that occur through industrial control to the monitoring platform (server);
• Server: The monitoring platform pushes alarm messages to the operation and maintenance personnel (client) through a series of alarm strategies;
• Client: After receiving the alarm notification, the operation and maintenance personnel go to the equipment point to maintain the equipment;
• Machine side: Equipment maintenance is completed, device status is updated and uploaded to the server.

After researching the products and user needs within the industry, the composition of the operation and maintenance fault alarm platform is divided into the following parts:
Fault data+fault monitoring+fault alarm+alarm handling+equipment health rating
1. Fault data
Fault data classification+fault data storage+fault data screening and filtering+productization of fault data warehouse

For different types of faults, targeted alarm strategies will be developed for triggering and pushing alarms.
Fault data storage: Based on the underlying design of the software and hardware of unmanned equipment, a set of storage fields that match the company’s business needs should be developed in advance, such as equipment number, fault name, fault code, fault start time, recovery time, duration, fault frequency, etc; As for the logic of data storage, due to the significant differences in business among different products, the author will not elaborate further;
Fault data screening and filtering: that is, manually filtering out faults that do not affect the normal transactions of unmanned equipment or interference faults generated by operation and maintenance personnel during cargo handling and maintenance;
The benefits are:

1. Reduce disruptive faults and focus on critical faults;
2. Reduce the attention cost of operation and maintenance personnel and improve work efficiency;

Data warehouse productization: Save each fault in a certain form to the productized warehouse for timely updates and maintenance in the future; Product design around data warehouse:

The display of faults is as follows: maintenance is carried out in the form of fault code+fault name+fault type+alarm indicator+urgency+liberation plan. The product function design supports:

1. Fault addition;
2. Fault inquiry;
3. Fault editing;
4. Addition of alarm indicators;

Of course, the addition of fault codes depends on the underlying design of the device at the software and hardware level. Interested students can conduct deeper research and learning, and I will not provide a detailed introduction;
The correspondence between “single fault” and “alarm indicators” will be detailed in the following fault alarms.
2. Fault monitoring
Based on actual business and needs, the author divides it into fault log monitoring and fault alarm monitoring;
Fault log monitoring: Real time monitoring and recording of a single device with a single fault as the minimum granularity;
Fault alarm monitoring: Using a single alarm task as a small granularity, record the real-time status and maintenance progress of a single device, and synchronize the alarm task and device status after maintenance by operation and maintenance personnel.

Fault alarm monitoring

By presenting a list in the form of “single device – single alarm”, a single alarm can contain multiple faults, and the status of the alarm task is ultimately used as the final key node for closed-loop monitoring. In terms of functional design, certain fields can be queried, filtered, and exported. At the same time, for the alarm task of a single unmanned device, detailed information within the task can be viewed, such as the time and personnel responsible for receiving the alarm task.
3. Alarm handling
The process in which frontline operation and maintenance personnel receive alarm messages through mobile clients and collect them until maintenance is completed at the device location is an important part of the fault alarm closed-loop;
Alarm processing is divided into: alarm message reception+alarm task collection+machine side fault maintenance and clearing.
The design principles for alarm processing are: clear message display, simple and easy to understand message content, convenient collection of alarm tasks, and convenient clearing of machine side alarms. The reason for placing the alarm clearing on the machine side is to prevent loopholes in personnel operations to a certain extent… (100 words omitted here);
Based on the above principles, carry out product design:
Firstly, for the mobile devices of operation and maintenance personnel

Provide a list of alarm tasks and priority sorting (priority sorting varies greatly depending on the business and strategy logic, so I will skip it here). At the same time, multiple alarm tasks under a single unmanned device can be freely accessed in the alarm details, and batch access is supported. After receiving the tasks, the information is synchronously accessed to the server.

After the equipment maintenance is completed, the operation and maintenance personnel enter the maintenance interface through the screen of the unmanned device, clear the corresponding alarm, and the alarm is completed. The completion status is synchronized to the backend server, and the entire operation and maintenance fault alarm loop is declared complete.