Abstract
Predictive maintenance (PM) involves the use of internet of things and machine learning techniques applied to machinery for remote monitoring of different variables to timely detect problems, before they require costly maintenance or generate customers’ complaints. Thus, it minimises the probability that the machine will break, extends its lifecycle and reduces the number of corrective actions. Despite the importance of PM and the growing implementation of Industry 4.0 technologies, a limited number of Italian machinery companies today includes PM systems in their products. Additionally, the topic has received little attention by literature. Consequently, there is a need to identify the barriers to the implementation of PM in the Italian machinery industry. Therefore, the aim of this research is to categorise the challenges to be considered when implementing a PM and propose a set of possible countermeasures. In doing so, the study reviews the existing literature on this topic and empirically explores three cases in the machinery industry. The results of the literature review show a list of barriers to PM implementation that can be related to the machinery industry. Then, the barriers are empirically validated, and inductively extended, and final set of countermeasures is proposed to overcome these challenges, in help of managers that are interested in adopting PM.