A leading automobile manufacturing industry that produces around 4 million cars per year, approached PREDICT to provide a solution that could predict failures up front, assist in efficient maintenance activity planning and, minimise the overall financial impact.
2.5 months Prognostics
Predicted the failure of linear motor 2.5 months in advance
Root Cause Identified
Deposits between the motor’s sliding boards led to the failure
16 hours of Production
Delayed maintenance resulted in 16 hours of downtime
The automobile manufacturing industry was looking for a solution that could
- Improve the operational availability of the assets
After a meticulous engineering study of the assets in the installation, a model was formulated to monitor the behaviour of the assets. This enabled in providing regular feedback to boost the operational capability of the assets.
- Reduce general expenses by replacing the right part at the right time
By using efficient predictive analytic algorithms, the remaining useful life of the asset is estimated so that maintenance could be performed with minimum intervention to the production activity.
- Optimise resources to reduce the preventive workload
By predicting the assets that would break in the future and, identifying the root causes that may lead to the asset breakdown, there was a substantial decrease in the preventive workload.
Analysing the engineering behind the asset and associating it with the historical data of 6 months from different sensors, PREDICT was able to model the behaviour of the system and identify the best indicators to predict the failures.
The solution when deployed in the client’s environment, was able to monitor newly installed equipment. By integrating inputs from the domain experts on the client’s side, PREDICT could further improve its solution in identifying the root causes.
Within a month of installation, PREDICT was able to predict a breakdown in one of the newly installed linear motors 2.5 months in advance. In addition, PREDICT could also determine that the deposits between the sliding boards of the linear motor would result in the failure of the linear motor.
However, maintenance was not performed on the linear motor and, the linear motor broke down within the time frame PREDICT estimated. The client could have saved 16 hours of production downtime if the maintenance was performed on time.