Challenge
An international calibration facility for oil and gas flowmeters wanted to significantly reduce the turnaround time for calibration cycles, improve the reproducibility of the process and reduce its reliance on experienced operators.
Solution
Over the course of a two-year development programme, large volumes of process data were analysed and machine learning models were developed to optimise the calibration process. The algorithms developed helped operators determine the optimal calibration settings and made the process more consistent and efficient.
Result
- Reduction in calibration time from 8 hours to 3.5 hours (a time saving of over 55 per cent)
- Improved reproducibility of calibration results
- Reduced reliance on operator experience
- Higher capacity and more efficient use of the calibration facility