contributions Design study success story . Highlighting matches and - PowerPoint PPT Presentation

Matthew Brehmer @mattbrehmer 
 UBC Jocelyn Ng @JocelynNg 
 Matches, Mismatches, and Methods: 
 EnerNOC Multiple-View Workflows for Energy Portfolio Analysis Kevin Tate EnerNOC paper & supplemental materials: 
 cs.ubc.ca/labs/imager/tr/2015/MatchesMismatchesMethods/ Tamara Munzner @tamaramunzner 
 UBC IEEE InfoVis October 2015


 contributions Design study success story . Highlighting matches and mismatches: 
 • task & data abstractions ←→ visual encoding & interaction design 
 • multiple concurrent time series 
 Addressing domain convention , familiarity & trust . 
 Reflecting on methods for visualization design studies. Brehmer et al. – InfoVis 2015 2 10-15

design process timeline 4 months full-time 3 months part-time 4 months part-time 2013 2014 2015 work visual production domain encoding development project task & data workflow analysis design by collaborator inception abstraction design Brehmer et al. – InfoVis 2015 3 10-15

outline: design process 1. analyzing the work domain • interviews with 9 energy workers 2. identifying data and task abstractions 3. visual encoding sandbox prototyping 4. eliciting feedback on vis. encoding designs 5. prototyping workflows 6. production development by collaborator Brehmer et al. – InfoVis 2015 4 10-15

43 summary slides energy workers’ skill sets, goals, activities work domain analysis existing tools workarounds Brehmer et al. – InfoVis 2015 5 10-15

21 energy worker artefacts Brehmer et al. – InfoVis 2015 6 10-15

outline: design process 1. analyzing the work domain 2. identifying data and task abstractions 3. visual encoding sandbox prototyping 4. eliciting feedback on vis. encoding designs 5. prototyping workflows 6. production development by collaborator Brehmer et al. – InfoVis 2015 7 10-15

data abstraction Brehmer et al. – InfoVis 2015 8 10-15 images: openclipart, pixabay, wikimedia commons

building portfolios Term Abstraction Example Building ID Unique categorical #123 Building area quantitative 450m 2 Location spatial 49.26º N, 123.25º W tag categorical “restaurant” Brehmer et al. – InfoVis 2015 9 10-15

raw time series data Term Abstraction Example Energy demand quantitative 200 kW Outdoor quantitative 18º C temperature Brehmer et al. – InfoVis 2015 10 10-15 images: pixabay, wikimedia commons

derived time series data Term Abstraction Example Consumption quantitative 800 kWh normalized Intensity 1.78 kWh / m 2 quantitative normalized % Savings 40% quantitative Rank ordinal 1st, 2nd, 3rd Brehmer et al. – InfoVis 2015 11 10-15

Brehmer & Munzner (2013), Actions Targets All Data Analyze Munzner (2014) Consume Trends Outliers Features Discover Present Enjoy Attributes Produce One Many Annotate Record Derive Dependency Correlation Similarity Distribution tag Extremes Search task abstraction Target known Target unknown Location Network Data Lookup Browse known Topology Location Locate Explore unknown Query Paths Identify Compare Summarize What? Spatial Data Why? Shape How? Brehmer et al. – InfoVis 2015 12 10-15