CityPulse has 120 public buildings, each described by 40 attributes: monthly energy use (12), floor area, age, building type, occupancy, heating type, insulation rating, and 20 sensor-derived indicators (temperature stability, night-time consumption share, peak-load ratios…). The energy office wants to find groups of similar buildings and buildings that behave unusually for their group.
Design a visual analysis: reductions (attributes and items), the views (at least three, linked), what each view hides, how you would validate that groups are meaningful, and how you would present the result to non-experts.