Using statistical models in industrial equipment

Abstract

Statistical methods are nowadays more and more useful in industrial engineering. From plant design reliability to equipment analysis, there is much to cover with statistical models in order to improve the efficiency of systems. At Sines refinery we used several approaches trying to relate several process variables with the vibration of critical equipment and to model time to failure using parametric and non-parametric models. Critical pumps are also a purpose of the study and we focus on a 2oo3 (two-out-of-three) structure in order to check if maintenance optimization or maintenance cost reduction is possible, and thus, it will be included the study of the operation of some rotary equipment.