Jun 08, 2023
No image
Completed
Locomotive failure prediction system
$100,000+
4-6 months
United States
6-9
Service categories
Service Lines
Artificial Intelligence
Domain focus
Transportation & Logistics
Programming language
Python
Frameworks
TensorFlow
Torch/PyTorch
Subcategories
Artificial Intelligence
Machine Learning
Challenge
Serokell has engineered a machine learning system for predicting electric locomotive engine failures for a rail transportation company. The goal was to create a predictive system that minimizes costly breakdowns and reduces maintenance downtime by providing precise, understandable predictions. During the project, we encountered several challenges pertaining to data quality, such as biases, anomalous data points, and gaps in the data.
Serokell has engineered a machine learning system for predicting electric locomotive engine failures for a rail transportation company. The goal was to create a predictive system that minimizes costly breakdowns and reduces maintenance downtime by providing precise, understandable predictions. During the project, we encountered several challenges pertaining to data quality, such as biases, anomalous data points, and gaps in the data.
Solution
The developed a solution based on the Internet of Things (IoT) and machine learning technologies. Specialized sensors were installed to gather data that is further processed through big data analysis. As a result, the system can accurately predict the time to failure.
The developed a solution based on the Internet of Things (IoT) and machine learning technologies. Specialized sensors were installed to gather data that is further processed through big data analysis. As a result, the system can accurately predict the time to failure.
Results
As a result, the system can accurately predict the time to failure up to 7 days ahead and can spare for the business partner the cost of expensive preparation by warning about the necessity of just-in-time maintenance.
As a result, the system can accurately predict the time to failure up to 7 days ahead and can spare for the business partner the cost of expensive preparation by warning about the necessity of just-in-time maintenance.