Artificial Intelligence Distributionally Robust Optimisation of Community Clinic Oxygen Concentrator Fleet Maintenance Scheduling Under Variable Compressor Failure Telemetry: A computed Healthcare Continuity Evaluation

Community clinic oxygen concentrator fleets fail through compressor degradation whose telemetry is sparse, delayed and heterogeneous across devices, so maintenance schedules optimised against a single empirical failure distribution can leave clinics without oxygen when the realised distribution shifts. the present analysis formulates fleet maintenance scheduling as a distributionally robust optimisation problem over an ambiguity set of compressor failure distributions calibrated to telemetry uncertainty, and evaluates the resulting schedules in a computed healthcare continuity study. The design is a before-after-control-impact evaluation with a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64), power estimated across 2000 replications per design cell for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis is methodological: no observational fleet data are used or claimed, and the reported quantities characterise the operating characteristics of the design rather than the performance of a deployed schedule. Under the declared variance structure, 56 clusters are required for 80% power at a 5 percentage point effect. The contribution is a formal bridge between distributionally robust scheduling and the cluster-level evaluation logic that healthcare continuity studies require, together with an explicit statement of the design conditions under which that bridge holds.