MerSci helps manufacturers rapidly integrate data from enterprise systems, operational sources, sensor networks, and external providers to power machine learning models that generate predictive insights. This addresses important goals such as inventory stock reduction, waste elimination due to quality defects, and generating economic value annually for manufacturers.

Client Issue
  • Several machines and robots are working on the lines in the factory and each contains several motors and cylinders, which produce signals stored as event log data every millisecond.

  • The factory wants to monitor the stability of all the signal and investigates whether there is any anomaly during the production.

  • Furthermore, the factory wants to cary on the anomaly detection model to predictive maintenance model and create an alert system for engineer.

MerSci Approach
  • Data Scientists at MerSci created a data driven solution by connecting the machines to the cloud services and built dashboards to monitor the signals for each machine.

  • Also, anomaly detection algorithms such as MAD and DBSCAN used to investigate the anomaly situation over the production lines and gives alert in the case of abnormal situation.

  • The anomaly detection results led the team to propose a predictive maintenance model that helps engineers and operators to take action in advance when it is necessary.

Value Created
  • The user-friendly and dynamic dashboards helped the factory to monitor signals in all the machines and robots on the line and check the stability of the machines over the each production cycle.

  • The factory receives alerts using the anomaly detection models and take immediate necessary action based on the alert and machine.

  • The engineers and operators are always one step ahead for maintenance activities based on the advices that they receive from the predictive maintenance models.

Take Your Next Step

Meet the MerSci team and find out what it's like to work with us,