CONTEXT DRIVEN MAINTENANCE: AN eMAINTENANCE APPROACH

All assets necessarily suffer wear and tear during operaƟon. PrognosƟcs can assess the current health of a system and predict its remaining life based on features capturing the gradual degradaƟon of its operaƟonal capabiliƟes. PrognosƟcs are criƟcal to improve safety, plan successful work, schedule maintenance, and reduce maintenance costs and down Ɵme. Prog nosis is a relaƟvely new area but has become an important part of CondiƟon‐based Maintenance (CBM) of systems. As there are many prognosƟc techniques, usage must be aƩuned to parƟcular applicaƟons. Broadly stated, prognosƟc methods are either data‐driven, rule based, or model‐base d. Each approach has advantages and disadvantages; conse‐ quently, they are oŌen combined in hybrid applicaƟons. A hybrid model can combine some or all model types; thus, more complete informaƟon can be gathered, leading to more accurate recogniƟon of the fault stat e. This approach is especially relevant in systems where the maintainer and operator know some of the failure mecha‐ nisms, but the sheer complexity of the assets precludes the development of a complete model‐based approach. The paper addresses the process of data aggregaƟon into a contextual awareness hybrid model to get RUL values within logical confidence intervals so that the life cycle of assets can be managed and opƟmised.