1 1 Developing a framework for standardisation High-Level Seminar on Streamlining Statistical production Zlatibor, Serbia 6-7 July 2011 Rune Gløersen IT.

1

1

Developing a framework for Developing a framework for standardisationstandardisation

High-Level Seminar on Streamlining Statistical production

Zlatibor, Serbia 6-7 July 2011

Rune GløersenIT DirectorStatistics Norway

2

Contents

• Preconditions for improved standardisation

• The characteristics of processes and data at NSIs

• Applicable standards for various business processes

• Governance

• Some international trends

3

Reasons for standardising statistical production

• Leaving stovepipes– Shift of focus from surveys and products to processes

• Introduction of quality frameworks– Coherence and comparability, quality assessments– Quality assurance and audits, risk reduction

• Improving efficiency– Internal interoperability, streamlining work processes, cost effectiveness

• Globalisation– International interoperability, comparability, benchmarking

• Content standardisation– Data and metadata standards– Best practise methods

• Technological standardisation– High-level architecture, standardise and reuse tools

4

4

QualityQuality

5

Model for Total Quality and Code of Practice

Use rs :Perc eived qualityNeeds and effec ts

Ins titutio nal E nviro nm e nt:Struc tural quality

S tatis tical O utp ut:Produc t quality

S tatis tical P ro ce sse s :Proc ess quality

The point of departure fors ys tem atic quality w ork isthe "us er needs "

The us ers dem and "produc t quality":R elevanc eAc c urac y and R eliabilityT im elines s and Punc tualityC oherenc e and C om parabilityAc c es s ibility and C larity

S tudy of proc es s es is a prec onditionfor im provem ents :

Sound MethodologyAppropriate s tatis tic al proc eduresNon-Exc es s ive Burden on R es pondentsC os t effec tivenes s

S truc tural c onditions provide a fram ew ork:P rofes s ional Independenc eMandate for D ata C ollec tionAdequac y of R es ourc esQ uality C om m itm entS tatis tic al C onfidentialityIm partiality and O bjec tivity

6

6

Cornerstones of standardisationCornerstones of standardisation

7

Cornerstones of standardisation and improved interoperability

Organisationalinteroperability

Technologicalinteroperability

Semanticalinteroperability

8

Enterprise ArchitectureCoherence and interoperability

Generic StatisticalBusiness Process Model

ICT-Architecture(Principles)

Generic StatisticalInformation Model

Best PracticeStatistical Methods

9

9

Business processesBusiness processes

10

10

GSBPM – leaving stove pipes

11

11

Data and semanticsData and semantics

12

Specifyneeds

Design Build Evaluate

Quality Management/Metadata Management

Process stages and data archiving

Data archiving spans the 4 main business processes,and comprises 4 steady states of the data life cycle

13

Classifi-cations

Interface

Master Metadata

Statistics production systems

DataDoc

Variables

StatisticalProducts

Input datadefinition

Users

Interface

About theStatistics

14

Specifyneeds

Design Build Evaluate

Quality Management/Metadata Management

Adopting standards

DDISDMX

?

15

15

ITIT

16

10 IT architecture principles

• IT-solutions must be built upon standard methods, a standard infrastructure and be in accordance with Statistics Norway’s business architecture

• IT-business alignment

• Open standards

• Our IT-solutions must be platform independent and component based, shared components must be used wherever possible

• It must be possible to create new IT-solutions by integrating existing and new functionality

• Our services must have clearly defined, technology-independent interfaces

• Distinguish between user interface, business logic and data management (layered approach)

• End-user systems must have uniform user interfaces

• Store once, reuse may times (avoid double storage)

• Data and metadata must be uniquely identifiable across systems

17

A layered and modularised model for a coherent, streamlined production system

Users

Monitor and Manage

Tools and services

Workflow

Data and metadata

18

18

GovernanceGovernance

19

Project Portfolio ManagementPrioritized and followed up by the Top Management

Project-Proposal

Decision Planning DecisionProjectExecution

Project-Directive

Assessment

Priority parametersScoreComments

Project-plan

Approved project plansDetailed requirementsAvailable resources

ReportsAssessment

20

Statistics NorwayOrganisation and Management

SLACommon Services

Appendix

SLADepartments

Appendix

Service Level AgreementsSystems Maintenance

• One SLA for each of the subject matter departments

• Approx. 400 IT systems are maintained for the 4 Statistics Production Departments. In addition approx.70 systems for data collection, administration and dissemination

• An SLA covering Common Services is under development

21

21

Some international trendsSome international trends

22

The diversity of users, needs and data flows

Public(re)use

Domain specificanalysis

Research andData Integration

Questionnaires

Data transfersRegisters

Common high level models, vocabulary etc

23

Maturity growth in e-Government

OrganisationalInteroperability

SemanticalInteroperabilitySource: www.semicolon.no

Analytical Framework for e-Government Interoperability

SharingKnowledge

Aligning WorkProcesses

Joining ValueCreation

AligningStrategies

Bilateral data exchange, semi automated,Technical specifications and standards

Share best practises, metadata specifications,Set up standards for technical systems and dataexchange

Common information models, process models and service catalogues, shared development costs

Legislation,Whatever

24

Common GenericIndustrial Statistics

GSBPM GSIM

Methods Technology

Statistical Concepts Information Concepts

Statistical HowTo Production HowTo

conc

eptu

alpr

actic

alIndustrializing Statistics

De-coupling content and technical standardisation

25

25

Thank you for your attention!Thank you for your attention!

Questions or comments…