This project used timestamped mission-log data to create a lightweight digital shadow of a collaborative robot picking operation. The objective was to understand where robots were spending time, identify the source of process delays, and evaluate the likely impact of a proposed operational change.

Challenge
The system operated across three primary stages: induction, picking, and drop-off. Performance issues were visible in the operation, but the available data was limited to event timestamps rather than a full system model or detailed sensor history.

Approach
- Built a simplified digital shadow using timestamped mission records.
- Tracked robot movement through induction, picking, and drop-off.
- Calculated mission time, processing time, queue time, and stage-level variability.
- Visualized robot flow and waiting behavior across the process.
- Identified the stage contributing most heavily to overall cycle time.
- Tested the expected impact of a revised process design.

Outcome
The digital shadow was developed and deployed in less than one day and quickly identified the drop-off station as the primary system bottleneck.

After the process was redesigned, the model was used to validate that the change reduced queueing, improved robot flow, and increased overall throughput. The project demonstrated how a lightweight model can support fast, practical decision-making even when only limited operational data is available.