Effective maintenance policies can support companies to deal with process interruptions and consequently, to prevent significant profit losses. Moreover, the proliferation of structured and unstructured data due to production plants validates the application of knowledge discovery in databases techniques to increase processes’ reliability. In this paper, an innovative maintenance policy is proposed. It aims at both predicting components breakages through association rule mining and determining the optimal set of components to repair in order to improve the overall plant’s reliability, under time and budget constraints. An experimental campaign is carried out on a real-life case study concerning an oil refinery plant. Finally, numerical results are discussed considering different blockage categories and number of components and by varying some significant input parameters.
A predictive association rule-based maintenance policy to minimize the probability of breakages: application to an oil refinery / Antomarioni, Sara; Pisacane, Ornella; Potena, Domenico; Bevilacqua, Maurizio; Ciarapica, Filippo Emanuele; Diamantini, Claudia. - In: INTERNATIONAL JOURNAL, ADVANCED MANUFACTURING TECHNOLOGY. - ISSN 0268-3768. - 105:9(2019), pp. 3661-3675. [10.1007/s00170-019-03822-y]
A predictive association rule-based maintenance policy to minimize the probability of breakages: application to an oil refinery
Antomarioni, Sara
;Pisacane, Ornella;Potena, Domenico;Bevilacqua, Maurizio;Ciarapica, Filippo Emanuele;Diamantini, Claudia
2019-01-01
Abstract
Effective maintenance policies can support companies to deal with process interruptions and consequently, to prevent significant profit losses. Moreover, the proliferation of structured and unstructured data due to production plants validates the application of knowledge discovery in databases techniques to increase processes’ reliability. In this paper, an innovative maintenance policy is proposed. It aims at both predicting components breakages through association rule mining and determining the optimal set of components to repair in order to improve the overall plant’s reliability, under time and budget constraints. An experimental campaign is carried out on a real-life case study concerning an oil refinery plant. Finally, numerical results are discussed considering different blockage categories and number of components and by varying some significant input parameters.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
