The old design showed throughput overview stats but, the users did not have any context for what these numbers mean, i.e. Is the throughput good to meet targets, Is it deviating from expected.
From my research, I was able to understand that the system showed multiple streams of data which was captured by AI Camera to show the throughput
I collaborated with System Architect AI engineers to understand what all data was captured by the AI system.
leverage the power of AI for better data visualization, I brought the idea to my PM to be able to filter through data to set trigger points for incoming data from the system. Once, the team was on board, I was able to propose the idea for building and Action Centre, which would automatically notify the users for anomalies in the assembly line
The process to understand and design the Action Centre involved looking at all data points on the platform, Understand how data was gathered by the AI system, look at how the APIs were pulling this data, Work with engineers to understand the overhead from their team to deploy Alerts based on set thresholds for the data.
Jumped into wireframing, the below screen shows the different interactions and changes made based on the review sessions and feedback received from the team.
We finally settled on the below designs for throughput overview and action center given the time constraint we were operating in, moving certain features to the future design sprint
.png)
The new redesigned throughput overview is split between two cards.
The Stats card which shows ongoing throughput, projected throughput and the downtime caused in the overall shift along with hourly breakdown to show deviation from target.
The Action Center shows Alerts based on High, Medium and Low priority, which can be filtered through. To understand what went wrong the users can go to the camera feed to observe exact instance that triggered the alert.
.png)
.png)
.png)
.png)
.png)
.png)
Go to Solution ↑