Scope of the workshop

About the workshop

Smart devices, electronic networks and constant generation of data on all aspects of life and the environment will become an integrative component of how our societies and economies will function. Most if not all data in the course of the third decade of the 21st century is expected to be "organic", i.e. by-products people's activities, systems and things, including billions of low-end and affordable smart devices connected to the internet i.e. Internet of things - IoT. In addition, the fourth industrial revolution (Industry 4.0) and the industrial IoT are transforming manufacturing operations, bringing smart technologies in the spotlight and automation to a higher level of integration of smart systems.

The IoT has the potential to offer added value to the production of official statistics. Statisticians have a new opportunity to produce official statistics that use an extended IoT data ecosystem.

The workshop will be oriented towards the needs of the European Statistical System, which strives to better serve users’ needs. It will also demonstrate the opportunity and necessity of leveraging smart statistics for use in decision-making.

Based on proofs-of-concepts or prototype applications, the presentations should shed light on the potential for the production of Trusted Smart Statistics.

Issues related to the following will be discussed:

  • use of smart technologies and smart systems for statistics,
  • citizen-science data using wearables and smart technologies,
  • smart services (e.g. smart farming) for smart statistics,
  • standards and methods for using the IoT for statistical purposes,
  • computational intelligence,
  • privacy-preserving architecture,
  • GDPR in the context of smart statistics
  • ...

The workshop is organised by Eurostat with the assistance of the Federal Statistical Office of Germany (DESTATIS), the Jožef Stefan Institute (Slovenia), and SOGETI Luxembourg S.A.


The topics to be addressed during the Workshop on Trusted Smart Statistics: policymaking in the age of the Internet of Things include:

  • Use of technologies embedded in smart systems to produce trusted smart statistics: The technological capacity of smart devices/systems has evolved,  increased and become more complex over the last fifty years. The latest generation of smart systems combine technical “intelligence” and cognitive functions, providing an interface between the virtual and the physical world, e.g. wearables measuring physical activity. We will examine:
    • How the data produced, communicated and processed for a specific purpose or a specific service, can be used for producing official statistics? (e.g. data from sensors that are produced for smart farming, smart cities, smart traffic, etc.)
    • To what extent, and under what conditions, a "specific purpose" data layer can be transformed to accommodate the production of statistics
    • The necessary conditions, prerequisites and standards that may be required to transform smart technologies’ data into information
    • The mathematical-statistical methods and IT tools that would allow statistical inference by combining design-based, model- or algorithm-assisted approaches
    • How statistical concepts can be (re)defined, adapted to data sources, data availability and data processing capacity, in order to use a common denominator to enable the production of cross-country harmonised and comparable official statistics?
  • Quality and trust: Within the community of official statistics, “trusted statistics” is defined in terms of ‘compliance with quality standards’. National quality assurance frameworks, statistical codes of practice and compliance with international standards, can generally be used as a measure for “trusted data”. Additionally, the notion of trusted statistics is strongly correlated to the validity and accuracy of output as well as the respect for the data subjects' privacy and the assurance of confidentiality. We will discuss the following:
    • In a world of rapidly spreading IoT, how can we ensure the principles of respecting a data subjects' privacy and protecting confidentiality?
    • In a future hyper-connected environment of smart devices and ever-changing smart technologies, what would be needed to define the quality of official statistics?
    • How can we embed into statistical production processes issues such as privacy by design, end-to-end security, auditable data life-cycles, secure multi-party computation and other privacy-preserving techniques?
    • What are the prerequisites for the ESS to offer trusted smart statistics as a service to users?
    • To what extent should we develop a "trustmark" for smart statistics based on IoT? (e.g. a label for consumers to decide which devices to choose to trust)
  • Use of algorithms: Beyond a rule-based set of instructions, algorithms may have to learn from data, adjust, and take decisions in the context of statistical operations. Therefore, which decisions can cognitive, learning and problem-solving algorithms take?
    • How can we ensure algorithmic transparency and accountability?
    • Who can verify that the relevant principles are respected?

Other topics could include the following:

  • Data, tools and techniques (quality, sustainability, reliability)
  • Organisational changes, future data analytic capabilities, extended involvement of stakeholders and subject matter experts, skills

Fields of application

The following areas in which smart statistics may be applied (proofs-of-concept, research projects, combination of data sources, etc.) may be presented and explored throughout the workshop:

  • Use of citizen science data through smart technologies
  • Smart cities, smart environment, connected vehicles
  • Smart farming
  • Industry 4.0 - Industrial data space

Target audience

Our target audience for the workshop are members of the European Statistical System who work in the area of big data from different perspectives (methodology, business, information technology).

However, representatives of users of official statistics (e.g. policy analysts, academic researchers and journalists) and outside specialists on data analytics or standardisation bodies (e.g. data scientists or others) may also be involved as presenters.

Through group discussions, the workshop will provide an opportunity for participants to meet one another, share experiences and identify opportunities for cross-border collaboration and developments. The output of the workshop should help reinforce a mutual understanding of the underlying concepts and determine future work of this type at ESS level.