Mejora del plan de mantenimiento predictivo de equipos dinámicos en plantas de síntesis química, adaptado a la Industria 4.0 y basado en valoración multiparamétrica de la condición funcional

Zusammenfassung

The intense globalization experienced during the last decades in world trade has offered productive company’s great opportunities for growth, however, the irruption at the beginning of the year 2020 of the health and economic crisis caused by COVID-19 has caused this perspective to falter, exacerbating the traditional problem inherent in globalization such as the great inter-company competitiveness promoted by the low manufacturing prices of emerging countries. Although this fact may hinder even the mere subsistence of western companies, economic progress is still possible, and it is necessary to abandon the traditional concept that industrial growth is due solely to economic stimulus activities aimed at the renewal of physical assets, adopting instead a much more beneficial objective for the company, which is to obtain the maximum use of the facilities already existing. Success in this task depends to a large extent on the implementation of an adequate maintenance plan, which must also be perfectly adapted to the new industrial trends promoted by the paradigm of "Industry 4.0", whose fundamental pillar is the massive application of new information technologies, in view of optimizing all the processes of the factory. This thesis focuses directly on this direction, presenting a structured set of methodological proposals of a general nature whose application is based on the integration of predictive maintenance information and plant production in order to reduce failures during processes and, therefore, maximize the chances of success of the maintenance plan. The first of the methodologies developed, it is proposed to summarize the data and predictive and diagnostic information of the machines evaluated in a single indicator of their functional state, easily understandable for all levels of the organization and through which to easily assess their "state of health". The indicator mentioned above, is obtained from the evaluation of one or more parameters representative of the condition of the machines, the second methodology elaborated, allows to establish, for any of these parameters of functional significance, the levels of severity most appropriate for each machine and, in addition, that these are dynamic and adaptable to each case. However, for most of the production equipment in the process plants, the measurement of the parameters used for predictive maintenance is done on a regular basis. Therefore, the key to the success of the above methodologies lies in a correct choice of period between measures. This is the objective of the following methodological proposal, the formulation of a method to standardize the criteria that allow to define and change dynamically and automatically the periods between the successive measurements of parameters of functional significance to be made on each machine, what we have called Temporal Interval Between Measures (ITEM). Finally, the last of the methodologies presented, although it applies classic techniques of predictive maintenance, is based on the advantages provided by the integration of maintenance and production data in the ERP of the company, which has shown that having perfectly interconnected management tools for this type of information, makes it possible to demonstrate relationships between apparently independent data, making it possible, therefore, to detect repetitive failure patterns to correct them.