Reliability and Maintenance strategies for the Gas Systems of the LHC experiments

The aim of this thesis is to analyse and improve, by a systematic approach, the maintenance strategies used for the gas systems of the CERN LHC experiment detectors. These systems are very complex, consisting of numerous racks and components that need to be continuously in run phase to provide the gas mixture to the detectors. Therefore, it is essential to establish a structured and robust approach, such as Reliability Centered Maintenance (RCM), to set a methodology with the ultimate goal of increasing the reliability and availability of these systems. In particular, the analysis focuses on the gas systems for the Resistive Plate Chambers (RPC) detectors, chosen as a reference. The strategy involved examining all past maintenance activities, recorded failures and interventions documented in the CERN Gas Team logbook, known as eLOGs. This approach allowed the extraction of valuable information on the technical history of these systems, highlighting critical issues in the past management of maintenance activities. For example, a significant difference between the ATLAS and CMS RPC entries is spotted, despite the systems are quite similar. This study aims to gather the necessary information to implement an industrial and widely used tool: Failure Modes, Effects and Criticality Analysis (FMECA). To apply this procedure, all mechanical components were listed, starting from those present in the P&IDs, and subdividing them module by module. For each component, one or more failure modes were identified along with their potential causes and effects on the module and on the system as a whole. At this point, the criticality risk indexes, frequency and severity, and the risk matrices, defined in the work methodology were used to rank the failures. This procedure allows the prioritization of the most critical events with dedicated preventive maintenance tasks or improving the system design through interventions to reduce the severity of failures. The analysis of the studied systems, for the definition of a working methodology, lead to some design improvements, both for the point of view of the maintenance and of the operations, that are considered for the implementation by the Gas Team. For example, for some critical components preventive maintenance tasks are scheduled, as well as some improvements in the control logic, like the addition of warnings before full stop alarms are triggered. Moreover, this study allowed the development of a general and systematic approach for the maintenance management that can be extended to the other systems.