Meridian plans to triple its fleet size and add sensors to new aircraft types over the next three years, and leadership wants the predictive-maintenance data architecture to keep working without a redesign each time the fleet grows. What should the PM ensure the data architecture is designed for?

Designing scalable data architectures for growing datasets means anticipating future volume and variety — new aircraft types, more sensors, more data sources — so the system can expand (e.g., via partitioning, horizontal scaling, flexible schemas) without requiring a full architectural rebuild each time. Meridian's stated three-year fleet growth plan is exactly the kind of forward-looking requirement the PM must factor into architecture decisions now. The fixed-capacity system sized only for the current fleet, to be rebuilt from scratch at each growth milestone, guarantees repeated disruptive rebuilds, which is costly and risks recurring project delays timed to each growth milestone. The design that assumes no new aircraft types or sensors will ever be added is directly contradicted by the stated business plan and would leave the architecture obsolete almost immediately. The architecture optimized purely for minimizing storage cost today, regardless of future data growth, optimizes for the wrong variable; minimizing today's cost while ignoring growth typically leads to a more expensive, disruptive re-architecture later, which is a worse total outcome. For the exam, scalability planning is the forward-looking counterpart to the 'construct pipelines optimized for artificial-intelligence workloads' enabler — one addresses today's workload shape, the other addresses tomorrow's growth.