The Need of Industry to Go FAIR

Abstract

The industry sector is a very large producer and consumer of data, and many companies traditionally focused on production or manufacturing are now relying on the analysis of large amounts of data to develop new products and services. As many of the data sources needed are distributed and outside the company, FAIR data will have a major impact, both by reducing the existing internal data silos and by enabling the efficient integration with external (public and commercial) data. Many companies are still in the early phases of internal data “FAIRification”, providing opportunities for SMEs and academics to apply and develop their expertise on FAIR data in collaborations and public-private partnerships. For a global Internet of FAIR Data & Services to thrive, also involving industry, professional tools and services are essential. FAIR metrics and certifications on individuals, data, organizations, and software, must ensure that data producers and consumers have independent quality metrics on their data. In this opinion article we reflect on some industry specific challenges of FAIR implementation to be dealt with when choices are made regarding “Industry GOing FAIR”.

Data Shepherding in Nanotechnology. The Initiation

“…To date, this is supported by the DMPTool (https://dmptool.org/, accessed on 1 May 2021), DMPOnline (https://dmponline.dcc.ac.uk/public_plans, accessed on 1 May 2021), and by inviting authors of exceptional DMPs to publish them in Journal collections to achieve machineactionable DMPs and to create a dynamic inventory of digital research-related outputs with shared benefits for diverse stakeholders [21]. While there are at the moment no protocols or standards to make nano-data FAIR, several efforts are underway [2]. If several projects reveal their actual processes towards FAIRification, such an act will accelerate the entire process and will narrow down the procedure into a more "Standardized way to make data FAIR in case where a universal template and ontology is missing".…”

“…As a specific example of the first approach, the genomics community has developed a curation workflow that uses standardized file formats for both metadata and raw DNA sequence data for submissions into standard repositories [12]. As for the second example, according to GO FAIR (https://www.go-fair.org/, accessed on 1 May 2021), the steps involved are as follows: (1) retrieve non-FAIR data, (2) analyze the retrieved data, (3) define the semantic model, (4) make data linkable, (5) assign license, (6) define metadata for the dataset, and ( 7) deploy/publish the FAIR data resource [22]. While the aforementioned steps are clear, we here provide a transparent case/paradigm wherein the first steps of retrieving data are actualized and recorded.…”

“…The challenges and technical barriers included in the process of FAIRification are already stressed in the literature [2,3,5,9]. Regarding technical barriers, the successful deployment of FAIR across the entire life cycle assessment of the NEPs requires standardized information architectures.…”

“…Data annotation is a time-consuming and resource-intensive process posing a crucial challenge to overcome [3]. Choosing the correct concept, especially for researchers that have little to no experience with ontology mapping, is challenging, and performing ontology mapping for a complete data set is an extensive task [2]. To facilitate this process, some initiatives are created in order to find a common vocabulary.…”

“…The current digital innovation challenge is to link emerging technologies by bridging datasets across domains to consumer needs and societal demands [1]. In academia, much like in industry, data are rarely leveraged beyond their original intended purpose [2]. Technologies that produce data which meet the principles of Findability, Accessibility, Interoperability and Reusability (FAIR) are fundamental enablers for digital transformation [3].…”