Fluid　contamination　is　one　of　the　main　reasons　for　the　wear　failure　and　the　related　downtime　in　a　hydraulic　power　system.　Filters　play　an　important　role　in　controlling　the　contamination　effectively,　increasing　the　reliability　of　the　system,　and　maintaining　the　system　economically.　Due　to　the　uncertainties　of　system　parameters,　the　complicated　relationship　among　components,　as　well　as　the　lack　of　effective　approach,　managing　filters　is　becoming　one　of　the　biggest　challenges　for　engineers　and　decision　makers.　In　this　study,　a　robust　interval-based　minimax-regret　analysis　(RIMA)　method　is　developed　for　the　filter　management　in　a　fluid　power　system　(FPS)　under　uncertainty.　The　RIMA　method　can　handle　the　uncertainties　existed　in　contaminant　ingressions　of　the　system　and　contaminant　holding　capacity　of　filters　without　making　assumption　on　probabilistic　distributions　for　random　variables.　Through　analyzing　the　system　cost　of　all　possible　filter　management　alternatives,　an　interval　element　regret　matrix　can　be　obtained,　which　enables　decision　makers　to　identify　the　optimal　filter　management　strategy　under　uncertainty.　The　results　of　a　case　study　indicate　that　the　reasonable　solutions　generated　can　help　decision　makers　understand　the　consequence　of　short-term　and　long-term　decisions,　identify　optimal　strategies　for　filter　allocation　and　selection　with　minimized　system-maintenance　cost　and　system-failure　risk.