BACK TO SCHEDULE REGULAR

Session 5

Standards and Interoperability

3 June 2026
11:00 – 12:30
ŠIBENIK VI SHOW ON MAP

Session chair
Remi Prual
Consultant in Management and in Quality, Knowledge, Information, Statistics, Data and Metadata Management and Governance, Estonia

Read more Read less

Remi Prual has been actively involved in advancing the quality of official statistics for several decades through numerous projects and initiatives. He served as Quality Manager and Head of the General Department at Statistics Estonia between 2004 and 2018 and has also been a Board Member of the Estonian Association for Quality for two consecutive terms. At the European level, Remi has contributed to the ESS Quality in Statistics Working Group, the ESS Steering Committee on Quality, and various task forces on metadata standards, quality reporting, fine-tuning of the QAF, as well as the EU CWPS presidency team. More recently, he has worked as an ESS peer reviewer, management consultant, strategic advisor, trainer, and expert with a dedicated focus on the valuation of data, metadata, statistics, information, and knowledge. Remi is certified as an expert in ISO 9001, EFQM, CAF, ITIL, LEAN, and several other quality management-related standards and approaches. He holds master’s degrees in both Information Sciences and Information Management.

Presentation title
Data Integration in Official Statistics: The SSO’s Experience with the National Interoperability Platform
PRESENTATION PAPER

Data integration brings both challenges and opportunities for NSIs.

Read more Read less

It demands strong governance, sophisticated methodologies, reliable and secure technological environments as well as coordinated work across different institutions. When supported by robust quality assurance mechanisms and modern interoperability platforms, data integration becomes a key enabler of the NSI’s digital modernisation and its capacity to respond to emerging data needs.

The National Interoperability Platform, operational since 2016, enables standardised and secure electronic data exchange across institutions by applying common rules for security, data structure and communication. Its framework covers legal, organisational, semantic and technical aspects, ensuring both technological connectivity and institutional alignment.

The SSO was among the first users, initially consuming data services such as employment information. It later expanded its role by registering secure web services that allow other institutions to transfer data directly to the SSO, including automated Central Population Register (CPR) data exchange introduced in 2021.

In 2025, the SSO introduced interoperability platform web services to support the electronic survey of students via the National E-Services Portal, with all submitted data stored securely on SSO premises. In the same year, the SSO modernised the collection of monthly employment data by replacing the outdated transfer with a secure, automated data-exchange web service registered on the interoperability platform. This upgrade enables the Employment Agency (EA) to transmit and store data directly to SSO databases in a timely and reliable manner.

Through this journey, the platform has enabled the SSO to automate data flows, replace manual and outdated transfer methods, improve data timeliness, strengthen data security by storing information directly on SSO premises and reduce administrative burden for partner institutions and citizens. Overall, interoperability has enhanced the efficiency, reliability and modernisation of SSO’s statistical production processes.

Looking forward, the national goal is to consolidate all institutional data transmissions on the interoperability platform, establishing a fully unified, secure and standardised system. Achieving this goal will maximise efficiency, improve data quality, strengthen confidentiality and create a coherent framework for data exchange across all public sector organisations. As one of the pioneer users, the SSO is continuously involved in this process, as a prerequisite for establishing a State Data Governance System suitable for official statistics use.

Main author / Presenter
Dejan Stankov
State Statistical Office of the Republic of North Macedonia (SSO)

Read more Read less

Dejan Stankov, PhD, has served as Director-General of the State Statistical Office of the Republic of North Macedonia since August 2024. He comes from the position of Senior Advisor Analyst in the Statistics Department at the National Bank. Dejan Stankov holds a PhD in economics and an MSc in monetary economics, and he is an Adjunct Professor at the University of American College Skopje. He has more than 15 years of experience in the field of official statistics in the Macedonian statistical system, including as a Deputy Director-General of the State Statistical Office (2014-2017). His professional career was enriched by a traineeship at Eurostat, as a young statistician. He has participated in numerous statistical conferences, seminars, workshops and working group expert meetings organised within the European Statistical System, as well as IMF training courses.

CO-AUTHOR:

Presentation title
Quality Assurance in Statistical Information Systems Driven by Metadata
PRESENTATION PAPER

The Czech Statistical Office has started its key improvements of statistical information system under the support of EU project National Recovery and Resilience Plan in 2024.

Read more Read less

The project is called SIS 5.0 and it combines and develops current metadata oriented system following GSBPM principles. The project is composed of separated functional modules that communicate by both internal and external service bus. The most visible changes occurred for respondents, where we built respondent portal, modern online forms, connection to companies´ information system and supported use of extended administrative data in order to assure the data and matadata quality. However, significant changes are hidden to users and represent key challenges to the Office. The switch from Oracle platform to PostgreSQL represents significant impact on the operating costs of the system. Complete reconstruction of internal applications used for daily routine statistical production also requires lots of training for the staff of the Office. The expectation of the management of the Offices is a modern, safe and flexible statistical information system that will be well accepted by both respondents and customers.

Main author / Presenter
Marek Rojíček
Czech Statistical Office (CZSO)

Read more Read less

Mr Marek Rojíček has been with the Czech Statistical Office since 2001, when he graduated with a master's degree from the Prague University of Economics and Business. His professional focus has been on macroeconomic statistics, particularly national accounts. He contributed to the implementation of European national accounting standards at the Czech Statistical Office ahead of the Czech Republic's accession to the European Union. His scientific work focuses on examining the role of economic structures in the transition process from cost-oriented to quality-oriented competitive advantage, as well as analysing the macroeconomic and structural impacts of globalization of economic flows. This was also the subject of his doctoral thesis. As part of his pedagogical activities, he participates in teaching macroeconomic analysis at the Faculty of Economics of the Prague University of Economics and Business. He has been the President of the Czech Statistical Office since March 2018.

CO-AUTHOR:

Presentation title
From Data Collection to Public Dissemination: Automating Data Pipelines and Ensuring Real-Time Data Quality
PRESENTATION PAPER

The Service de coordination de la recherche et de l’innovation pédagogiques et technologiques (SCRIPT) is a key entity under Luxembourg’s Ministry of National Education, Childhood, and Youth.

Read more Read less

Its mission is to drive pedagogical and technological innovation, coordinate educational initiatives, and ensure quality assurance across the national education system. In this work, we focus on the redesign of an end-to-end data pipeline for the SCRIPT. The pipeline aims to automate three critical processes: data ingestion, quality control, and statistical dissemination to the public through interactive dashboards. This project seeks to enhance data accessibility, transparency, and efficiency for stakeholders in Luxembourg’s education sector. The legacy system relied on manual procedures, including twice-yearly manual data ingestion from multiple data sources, identification and correction of data quality issues, and manual publication of results. These manual processes limited scalability and timeliness. The proposed system is characterized by an architecture in which data are ingested daily into a data warehouse, automatically transformed, validated, and disseminated. Potential data quality issues are automatically signaled to the data engineering team of the institution, which can then take remedial actions to clean the data. Data dissemination dashboards are directly connected to the data warehouse layer of the system, guaranteeing that the published indicators are coherent with the most recent data available. One contribution of this work concerns the automatic identification of data quality issues within the data warehouse. Building on recent research on data identification and cleaning techniques using large language models, our work modifies the framework introduced by Zhang et al. (Data Cleaning Using Large Language Models, 2025) to adapt it to contexts where strict data privacy constraints are in place. In particular, the framework has been adapted to work with locally deployed large language models that can be easily executed on commodity hardware, such as a standard laptop. This approach avoids reliance on external cloud providers, making it particularly suitable for public institutions operating under strict regulatory and confidentiality constraints. Overall, the proposed solution improves dissemination timeliness and transparency, reduces operational costs, and increases data reliability. By combining automation, modern data engineering practices, and privacy-friendly LLM-based techniques, this work introduces a practical and scalable approach to improving data quality and the frequency of statistical publications.

Main author / Presenter
Samuel Souque
Ministry of Education, Children and Youth (MENJE), Luxembourg

Read more Read less

Samuel Souque is a Data Engineer for the statistical division of the Service de Coordination de la Recherche et de l’Innovation pédagogiques et technologiques (SCRIPT). His role is to develop and maintain the data warehouse supporting the division's activities in research and innovation regarding the Luxembourgish educational system.

CO-AUTHORS: