摘要：This article aims at solving a problem raised by the paradigms of Industry 4.0 and certain manufacturing companies with a consequent number of production lines. The main problem is about the deployment of predictive maintenance of production lines on the scale of a large size manufacturing company. Indeed, this kind of manufacturing plant is composed of several heterogeneous production lines that are composed of many components. Therefore, to deploy generically predictive maintenance in this context, it is necessary to have a clear and structured information system and a standardized automation process. It is also needed to prioritize the analysis and the creation of models for the more critical components. Our work is in the early stage and aims at proposing a complete methodology that allows deploying generically predictive maintenance of production lines on large-size manufactories. The methodology will use maximum technological tools of the industry 4.0 to ensure digital continuity during the production line's lifecycle. Due to this early stage of work, after a prompt highlight of challenges raised by industry 4.0, this paper will focus on the constraints of the deployment of predictive maintenance. The need for standardization of automation and the proposal of a deployment method built on the analysis of the production process is illustrated by our first results.
