Playful Probes for Design Interaction with Machine Learning: A Tool for Aircraft Condition-Based Maintenance Planning and Visualisation

Aircraft maintenance is a complex domain where designing new systems that include Machine Learning (ML) algorithms can become challenge. In the context of tool for Condition-Based Maintenance (CBM) in aircraft planning, this case study addresses (1) use Playful Probing approach to obtain insights allow understanding how design interaction with ML algorithms, (2) integration Reinforcement (RL) agent Human–AI collaboration planning and (3) visualisation CBM indicators. Using science research approach, we designed Probe protocol materials, evaluated results by running participatory workshop. Our main contribution show elicit ideas practices estimation tools RL agent. Through workshop participants’ observation, which they played artefacts, Probes favour elicitation user requirements aid planner reliable plan turn possible understand represent indicators visualise them through trajectory prediction.