United Nations Economic Commission for Europe Statistical Division Exploring the relationship between DDI, SDMX and the Generic Statistical Business Process Model Steven Vale, UNECE

Steven Vale - UNECE Statistical Division Slide 2 Contents  Introducing the GSBPM  Links to DDI and SDMX  Combining Standards to Industrialize Official Statistics  Conclusions

Steven Vale - UNECE Statistical Division Slide 3 GSBPM – The Background  Statistical production has traditionally been organised by topic, e.g. transport, trade, …  Financial pressures are encouraging new ways of thinking  Some statistical organisations are moving towards a process-based approach  Others are considering a matrix approach

Steven Vale - UNECE Statistical Division Slide 4

Steven Vale - UNECE Statistical Division Slide 5 Which processes?  Defining and modelling processes in statistical organisations started at least 10 years ago “Statistical value chain” “Survey life-cycle” “Statistical process cycle” “Business process model”

Steven Vale - UNECE Statistical Division Slide 6 Which processes?  Defining and modelling processes in statistical organisations started at least 10 years ago “Statistical value chain”X “Survey life-cycle”X “Statistical process cycle”X “Business process model”X Generic Statistical Business Process Model

Steven Vale - UNECE Statistical Division Slide 7 Why do we need a model?  To define and describe statistical processes in a coherent way  To standardize process terminology  To compare / benchmark processes within and between organisations  To identify synergies between processes  To inform decisions on systems architectures and organisation of resources

Steven Vale - UNECE Statistical Division Slide 8 Applicability  All activities undertaken by producers of official statistics which result in data outputs  National and international statistical organisations  Independent of data source, can be used for: Surveys / censuses Administrative sources / register-based statistics Mixed sources

Steven Vale - UNECE Statistical Division Slide 9 Structure of the GSBPM Process Phases Sub- processes (Descriptions)

Steven Vale - UNECE Statistical Division Slide 10 Key features  Not a linear model  Sub-processes do not have to be followed in a strict order  It is a matrix, through which there are many possible paths, including iterative loops within and between phases  Some iterations of a regular process may skip certain sub-processes

Steven Vale - UNECE Statistical Division Slide 12 Links to DDI and SDMX  SDMX standards refer to business processes, but do not have a model  DDI has the Combined Life Cycle Model

Steven Vale - UNECE Statistical Division Slide 13 Generic Statistical Business Process Model DDI 3.0 Combined Life Cycle Model 1 Specify NeedsStudy Concept Repurposing (part) 2 Design 3 Build 4 CollectData Collection 5 ProcessData Processing (mostly) Repurposing (part) 6 AnalyseData Discovery Data Analysis Data Processing (part) 7 DisseminateData Distribution 8 ArchiveData Archiving 9 Evaluate Quality Management Metadata Management

Steven Vale - UNECE Statistical Division Slide 14 Combining standards  “Industrialization” of statistical production  GSBPM provides the framework, but more is needed: Information model Data / metadata formats Modular architecture Shared development of tools and methods

Steven Vale - UNECE Statistical Division Slide 15 Data / metadata formats  SDMX? Suitable for micro-data??  DDI? Suitable for exchange of statistical aggregates?  Combined approach? Trial by Australian Bureau of Statistics

Steven Vale - UNECE Statistical Division Slide 16 Combining standards?

Steven Vale - UNECE Statistical Division Slide 17 Conclusions  Too early to say for definite, but: No single standard meets all requirements Combine standards?  GSBPM  DDI  SDMX  Others?  More work needed!

Steven Vale - UNECE Statistical Division Slide 18 Questions and Comments? METIS Workshop on GSBPM and related standards, 5-7 October 2011, Geneva DDI community welcome!