Industry: Maritime & Offshore
Tech Stack: Python & React
UPS battery predictive maintenance​
​
Ampelmann has encountered significant operational challenges due to "Code Orange" alarms. Our predictive model forecasts 22% of potential failures three days in advance and 18% fourteen days out, maintaining a low false positive rate of 8%.
​
This initiative has shown a significant impact, with anticipated annual savings of €46,600 and an impressive first-year ROI of 366%. We plan to enhance this model by incorporating more sensor data and broadening its use to other systems.  
​​
#MaritimeInnovation #PredictiveMaintenance #AdvancedAnalytics #DigitalTransformation
Temperature sensor optimization
​
Failures in water temperature sensors can trigger costly DTE shutdowns. We've applied AI-driven modeling to anticipate these failures and identify data patterns.
The model is effective during inactive phases, and we are focused on improving its accuracy during active operations. Our future efforts will integrate additional sensors, enrich the training data, and tackle systemic issues like external influences on cabinet temperatures.
Looking forward
​​​
We are eager to implement these solutions and examine further enhancements in areas like pressure, cylinder position, and nitrogen pressure systems. This partnership has already provided valuable insights into Ampelmann’s operations, fostering more informed maintenance decisions and facilitating knowledge exchange among teams.
​
value
We delivered a comprehensive predictive maintenance solution that leverages advanced analytics to minimize unplanned downtime. Our data-driven insights enabled Ampelmann to forecast potential failures in UPS batteries and temperature sensors well in advance, reducing costly operational disruptions.
Ultimately, our partnership empowered Ampelmann to make proactive, informed decisions that strengthened their maritime operations.