Reliability Assessment of a Packaging Automatic Machine by Accelerated Life Testing Approach

Abstract

Industrial competitiveness in innovation, the time of the market introduction of new machines and the level of reliability requested implies that the strategies for the development of products must be more and more efficient. In particular, researchers and practitioners are looking for methods to evaluate the reliability, as cheap as possible, knowing that systems are more and more reliable. This paper presents a reliability assessment procedure applied to a mechanical component of an automatic machine for packaging using the accelerated test approach. The general log-linear (GLL) model is combined based on a relationship between a number strains, in particular mechanical and time based. The complete Accelerated Life Testing - ALT approach is presented by using Weibull distribution and Maximum Likelihood verifying method. A test plan is proposed to estimate the unknown parameters of accelerated life models. Using the proposed ALT model, the reliability function of the component is evaluated and then compared with data from the field collected by customers referring to 8 years of real work on a fleet of automatic packaging machines.The results confirm that the assessment method through ALT is effective for lifetime prediction with shorter test times, and for the same reason it can improve the design process of automatic packaging machines.