The Generic Statistical Business Process Model and its Implementation in Practice Steven Vale, UNECE

Contents Introducing the GSBPM Links to other standards Further development of the GSBPM Implementation in practice

practical conceptual GSBPM Statistical Concepts Information Concepts GSIM Common Generic Industrial Statistics Methods Technology practical Statistical HowTo Production HowTo

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 The traditional, or “stove-pipe” way of producing statistics was based on one survey for one output, often with custom-made processing. This is now seen as too expensive, inflexible and inefficient. Whilst some organisations are re-organising based on common processes, others, such as Statistics Sweden, are considering a matrix approach.

So statistical production can be organised in subject-matter stove-pipes (the green columns), or by processes (the pink rows), or in a matrix.

Terminology 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”

Generic Statistical Business Process Model Terminology Defining and mapping business 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

Why do we need a model? To define and describe statistical processes in a coherent way To standardize process terminology To compare and benchmark processes within and between organisations To identify synergies between processes To inform decisions on systems architectures and organisation of resources 8

Developing the GSBPM Developed by the UNECE Steering Group on Statistical Metadata (METIS) Based on the business process model developed by Statistics New Zealand Three rounds of comments made the terminology and descriptions more generic Adopted in April 2009 9

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

Structure of the GSBPM Process Phases Sub-processes (Descriptions) The GSBPM has several levels, starting with the production process itself. This is divided into 9 phases: Specify needs, Design, Collect, Process, Analyse, Disseminate, Archive, Evaluate Each phase is divided into a number of sub-processes represented by the pink boxes. Further, more detailed, levels can be added for national implementations, but would not be sufficiently generic to be included in the international model.

Structure of the GSBPM (2) National implementations may need additional levels Over-arching processes Quality management Metadata management Statistical framework management Statistical programme management ........ (8 more – see paper)

Not a linear model Key features 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

This is an example of a statistical production process moving through the model. You can see that it does not follow all sub-processes. It has an iterative loop between sub-processes in the Process and Analyse phases, and when it reaches the end, it does not go right back to the start. For regular processes, it is not necessary to Specify needs, Design and Build systems each time the process runs.

Links to other standards SDMX standards refer to business processes, but do not have a model DDI has the Combined Life Cycle Model DDI = Data Documentation Initiative, a standard for using and archiving large micro-data sets

Combining standards? The GSBPM could provide a framework for managing interfaces between data and metadata standards. For example, there is some discussion at the moment about using DDI and SDMX together in statistical production processes that start with microdata and end with aggregates. DDI was originally developed for managing microdata, and SDMX for disseminating and exchanging aggregates. Rather than trying to make one standard do everything, it might be better to use the relative strengths of both.

Functionality stretched too far? Here is an example of what happens when too much functionality is added. The complexity means that even the original use becomes difficult!

Further development of the GSBPM UNECE Task Force No change to model (for at least 2 years) 5 themes: National Implementations of the GSBPM Communication resources Metadata flows within the GSBPM GSBPM and process quality management Other groups using the GSBPM as a framework for their activities The Task Force on the Further Development of the GSBPM identified these 5 priority areas, and are developing materials and information on each of these, for presentation at a Workshop in October.

Workshop on Statistical Metadata Theme: Implementing the GSBPM and combining metadata standards Where: Geneva When: 5-7 October 2011 More information: UNECE website - www.unece.org/stats/documents/2011.10.metis.htm All welcome! And here is the information about that Workshop. Invitations have been sent to all statistical organisations in the UNECE region, and you would be very welcome to participate. For more information, please see the web site.

Implementation 30+ countries have adopted the GSBPM or national versions as a framework to describe statistical production Also used for: Quality management Cost allocation Time recording “Classification” of IT systems

Statistics Sweden

Czech Republic

Republic of Korea - KSBPM Governance Statistics-based policy management Statistical policy management Quality management Statistical coordination Planning Data collection Dissemination Design Data processing Archive Implementation Analysis Evaluation Support for the production quality Quality check in each production step Management of statistical production Support for population-related information Support for sampling design Support for enumeration districts and maps for production Production process pool Sharing of statistical business knowledge Metadata use Help desk

Questions and Comments? steven.vale@unece.org www.unece.org/stats/gsbpm