Title: Processorientated statistical production
1Processorientated statistical production
- IAOS Conference, October 16, 2008
- Åke Bruhn, Director, Process Dept,
- Statistics Sweden
2Content
- The journey to develop quality and reduce costs
- The normal situation today the situation
tomorrow - Which problem to solve
- Some results at Statistics Sweden
- Some lesson learned so far at Statistics Sweden
3The journey to develop quality and efficiency
TQM Structured quality work, the customer in
focus correct from the beginning
Change in organisation Input-thruput-output
Process investigation many different production
processes and/or production methods, tools,
systems to solve the same task
4The normal situation today each product has its
own processes
Products
5The situation tomorrow processes and products
in cooperation
Products
6Which problem to solve?
- Make the statistical production more effective
and reduce the cost with the same or higher
quality - The customers needs are not always in focus
- Unclear rolls and responsibility between the
product-owner, the process-owner and the producer - The steering of the today's processes is to weak
- The development of the today's processes is
unstructured and not enough standardized
7More problem to solve
- Expensive to develop, maintain and document many
systems and tools - Difficult to implement new and better methods and
tools and focus on competence development - Many systems result in different quality
- Development work is depending on the competence
in each product - The resources of the NSI are not used in an
effective way - Sweden The LOTTA project was established in
spring 2006
8Results from the LOTTA-project and the ongoing
work in the new line-organization
- New organisation established 1 January 2008
Process department, RD department - A common process map to follow for all
statistical products
9The Statistical Process
Evaluate and Feed back 8
Establish needs 1
Design and Plan 2
Create and Test 3
Collect 4
Process 5
Analyse 6
Disseminate and
Communicate 7
Prepare dissemination 7.1
Classifu and code micro data 5.1
Produce frame and register population 4.1
Design output 2.1
Create Measurement instrument 3.1
Produce statistical output 6.1
Identify information needs and availability 1.1
identify clients 1.2
Produce final output 7.2
Edit Micro data 5.2
Create and adjust Prduction tools 3.2
Edit macro data 6.2
Design frame, registerpopulation and sample 2.2
Draw sample 4.2
Establish client contact 1.3
Disseminate final output to client 7.3
Impute for non-response 5.3
Carry out disclosure control 6.3
Prepare Data collection 4.3
Designdata collection 2.3
Create production flow 3.3
Establish Information need 1.4
Complement micro data 5.4
Carry out Data collection 4.4
Designprocessing 2.4
Test measurement instrument 3.4
Establish final observation register 6.4
Communicate Final output 7.4
Negotiate and agree 1.5
Calculate weights 5.5
Transfer and store data electronically 4.5
Designanalysis 2.5
Test production tool and production flow 3.5
Interpret and explain 6.5
Select and archive 7.5
Designa dissemination and communication 2.6
Carry out Pilot study 3.6
Establish contents for dissemination
and communication 6.6
Designproduction flow 2.7
Implement production flow 3.7
Note Informal translation
Plan production flow 2.8
Support and Infrastructure
10More results from the LOTTA-project and the
ongoing work in the new line-organisation
- Processorientated thinking
- Structured process development organisation
- New standardised methods and tools are decided
and introduced for all processes - Quality management system (EFQM, SixSigma, ISO 20
252) - Structured competence development
11Some lesson learned so far
- Strengthen in our thoughts that this is the best
way to reduce cost and secure quality run
smarter not quicker - Take time and a lot of efforts to change
staff-members thinking, attitude and culture - Take time to develop processes and show good
examples
12Thank you for your attention!Questions and/or
comments?
- e-mail ake.bruhn_at_scb.se