The curiously acronymized ‘Reasonable Ontology Templates’ (OTTR) organization* continues the somewhat Norwegian predilection for applying semantic web technology to industrial contexts including oil and gas. The inaugural OTTR get-together was held virtually in January 2021. DNV’s Johan Klüwer presented work done in the READI* Joint industry project to apply the OTTR approach to interpret an ‘existing’ industrial vocabulary, the CFIHOS RDL and data dictionary. Klüwer has leveraged the venerable ISO-15926-14 upper ontology to ‘enrich’ CFIHOS’ tabular-format (CSV) data such that it is ‘ready for dissemination’ as linked data.
Another presentation, from Melinda Hodkiewicz, (University of Western Australia), covered ‘Digitalization and reasoning over engineering textual data stored in spreadsheet tables’ presenting the ‘case for OTTR’. Hodkiewicz addresses a very general problem in how engineering data is handled in the real world where a ‘vast amount of engineering text data is stored in spreadsheet tables’ and is therefore ‘not machine-interpretable’. Using an example from the IEC60812 standard for failure mode and effects analysis (FMEA), Hodkiewicz showed how knowledge captured in the FMEA tables can be extracted to support quality control and re-use. The result recalls earlier work on ISO 15926 where the tabular data is transformed into a spaghetti-like graph of objects and relationships amenable to semantic-style processing. The work was originally presented at the 2020 IFAC Workshop on Advanced Maintenance Engineering, Services and Technologies, AMEST 2020.
Comment: We have been tracking the ups and downs of semantic technology in oil and gas for over 20 years. In a nutshell, the early work on ISO 15926 proved too complex for the common mortals and a few years ago, CFIHOS, the capital facilities information handover standard was proposed by the Netherlands-based USPI-NL to simplify ISO 15925 by reducing its essentials to a set of spreadsheets. It is therefore rather curious that OTTR is now transforming CFIHOS’s spreadsheets back to the original semantic/linked data paradigm.
* OTTR is described as ‘a language with supporting tools for representing and instantiating RDF graph and OWL ontology modelling patterns’. The main contributor is Martin Skjaeveland from Norway’s Sirius Labs.
*READI Joint Industry Project (2018-) that targets digitalization of requirements for the Oil and Gas on the Norwegian Continental Shelf.
Presentations are available on the OTTR event website.
Extra, extra ... Sirius continues with its valiant attempts to find a use for semantic technologies with recent announcement of work, ‘blending AI and semantics’ in the Bosch-supported ‘Scaling Industrial AI in 4 dimensions’ SIndAIS4 project. One use case is semantic oil and gas data management. This is to combine information from wellbores, seismic investigations, and general geological knowledge to ‘assess what types of rock are in the reservoir and intersect the wellbore’ and to explore ‘semantic approaches to improve data integration and analytics’. OSDU brace yourselves!
This article originally appeared in Oil IT Journal 2021 Issue # 5.
For more information or to comment on this topic email here.