k-means algorithm. [1

To be successful in the digital era and advanced industries, the maintenance and optimal use of automatic equipment and machines are significant. Thus, the role of instrumentation equipment for correct measurement of the sensitive parameters in tools appears. Every organization needs high-precision measuring instruments to maintain its production quality. Besides, maintaining precision in measuring equipment requires controlling it by repeating its calibration. Accurate prediction of recalibration is significant because a short calibration interval increases the stopping time in the production line and the depreciation of measuring equipment. As a result, the increase in the stopping time can increase measurement uncertainties, causing other problems, such as quality loss in the production line. The present study aims to develop a method for timing the calibration of instrumentation equipment using Failure Modes and Effects Analysis (FMEA) and reliability, using the Risk Priority Number (RPN) and Reliability-Centered Maintenance (RCM) and via self-organizing map (SOM) neural network clustering. This method was implemented for 220 pieces of instrumentation equipment in the water supply system of Isfahan Zob Ahan Co. The research results show that this clustering leads to the change of calibration intervals and cost reduction in this part of Isfahan Zob Ahan Co.