Home/Tech Blog/O&M and malfunctions/After the device data is on the cloud, how should the operation board be designed?
O&M and malfunctions

After the device data is on the cloud, how should the operation board be designed?

From 37 indicators to 8, this is the pit we have stepped on on 2000 devices.

2020-11-021 min read395 views18 shares
After the device data is on the cloud, how should the operation board be designed?

Define roles before setting indicators

The store manager looks at restocking and malfunctions, the regional manager looks at online rate and sales per square meter, and the engineering department looks at the distribution of alerts. There are three sets of dashboards for three roles.

Eight core indicators

  • Device online rate
  • Average daily number of shipments
  • Number of out-of-stock aisles
  • Average fault recovery time
  • Daily turnover per machine
  • Restocking fulfillment rate
  • Payment success rate
  • Average order value

Alert noise reduction

Alerts with the same root cause are aggregated to avoid dozens of duplicate notifications being generated from a single network outage.

Refresh frequency

A 5-minute refresh interval is sufficient for the operations dashboard. Metrics with high real-time requirements should be placed in the alert channel, and there is no need to overcrowd the dashboard with all of them.

Need a custom solution for your project?

From terminal selection, protocol docking to deployment delivery, the engineering team can provide complete proposals and quotations.