Gutenbrain, AI powered information extraction for industrial documentation

The Industry sector can’t ignore the impacts on its business models brought by the digitalization of all aspects of human civilization and the accelerated transition to a knowledge economy. 

Artificial Intelligence is part of a multipronged revolution sweeping the Industry as it further adopts smarter technologies and more knowledge-intensive activities. 

Gutenbrain, an AI-powered solution for processing industrial documentation, paves the road to a more technology-savvy Industry that is gearing up to be better equipped to meet tougher targets and face today’s pressing challenges. 

A collaboration between Actemium Oil & Gas and Axians to reinvent Maintenance & Inspection Engineering 

In the Oil & Gas Upstream industry, the lifecycle of any new offshore or onshore asset starts with the Engineering, Procurement & Construction (EPC) phase. 

After the start-up of the facilities, ACTEMIUM intervenes for maintenance activities during operations. 

In addition, ACTEMIUM is well known to have the expertise to mobilize Project Teams during the EPC Phase, in order to prepare all the data required to be loaded in the Client’s Maintenance Management System. 

So, when the plant starts, all information is available, allowing the Client to achieve a smooth transition to the operational phase. 

This process is typically executed in the context of a Maintenance & Inspection engineering contract (also called MIEC). 

ACTEMIUM discussed this challenge with the LEONARD team, to be able to process all these documents with the support of an AI-powered solution. 

ACTEMIUM knew what the inputs were and what they wanted to achieve, but they didn’t have the data science and software engineering expertise to produce it. 

AXIANS Portugal was willing to develop the AI competency in their Data Driven Business Unit, so LEONARD engaged both ACTEMIUM and AXIANS Portugal and the GUTENBRaiN relationship started. 

 

Disrupting a manual and painstaking process 

 The first activity of a MIEC is to create the asset register, which is the hierarchy of all the equipment composing the facility (motors, valves, pumps, etc.). 

This is the very first mandatory milestone of a MIEC to be able to move on to other activities such as criticality studies (to define the criticality of each equipment), or the definition of the spare parts to be procured and put in stock for later use. 

Finally, ACTEMIUM defines the maintenance plans, procedures and manuals, and link them to the asset register. 

Coming back to the asset register creation, the process performed by ACTEMIUM’s team is to gather all documentation coming from the EPC contractor, manufacturers and suppliers. Each of these thousands of documents must be searched manually to identify all the tag identifiers of the equipment, group them by system and sort them as per their parent / child relationship. 

When there is a new revision of a document, it must be reprocessed to make sure there is no impact on the asset register. 

Currently, this activity is very heavy, time consuming and due to its inefficiencies represents an overhead to the global project. Information is hard to find, and due to the manual intensive process sometimes leads to errors. 

 

Gutenbrain: AI-powered information extraction for industrial documentation 

ACTEMIUM entrusted AXIANS Portugal to strengthen their value proposition in the maintenance engineering industry through innovation in AI. Leveraging ACTEMIUM’s business expertise, AXIANS Portugal developed, with the help of LEONARD, an innovative AI-powered solution to address the typical pain points of a traditional, manually intensive MIEC workflow and provide a smart and innovative answer for maintenance engineering: Gutenbrain! 

 

The first challenge: document ingestion, the need to extract the embedded data in engineering files. A typical MIEC project is composed of thousands of dispersed documents, from disparate sources and different file formats that ultimately need to comply with the project’s specifications. Gutenbrain can process large amounts of data from various file formats into a managed database. Complementarily, these documents/files can include text, tabular data and technical drawings, sometimes stored in legacy formats. Since this is not directly usable, Gutenbrain employs industry-leading Vision AI to analyse the contents with high prediction accuracy. 

The second challenge: extract asset characteristics and other relevant information from files to assist the team when performing the previously described activities. Gutenbrain does this by applying powerful pattern matching algorithms that can detect and extract all relevant assets from these files, to produce a valuable and readily accessible knowledge base that allows teams to quickly locate mission-critical information. 

The third challenge: tackle the complexity of managing multiple changes and updates in the project documentation throughout the process. Gutenbrain can automatically re-processes any document that is updated and signal the changes between different revisions or versions of the same document. This allows teams to quickly understand variances and make more informed decisions when significant readjustments are detected, thus preventing potential costly business mistakes. 

All the above features are consolidated in a user-friendly digital engineering interface, based on state-of-the-art web standards, allowing teams to manage project assets efficiently, augment and enrich the knowledge base and produce reporting deliverables seamlessly. 

Faster, better and stronger Maintenance & Inspection Engineering 

 What is Gutenbrain’s mission?  Bring to all stakeholders in the MIEC project lifecycle streamlined and collaborative business workflows, boost workforce productivity by reducing manual, error-prone and tedious tasks when compiling the project’s assets repository, enhance maintenance and inspection activities, intelligently prioritize critical systems, and produce relevant information to improve maintenance operations. 

In addition to boosting productivity and quality, this innovative AI-augmented workflow becomes a strong differentiator for ACTEMIUM in the MIEC segment, strengthening their value proposition and offering them a unique position in the market. 

The fructuous relationship between ACTEMIUM and AXIANS Portugal created unprecedent synergies that culminated in the initial release of Gutenbrain, the AI-enabled solution for intelligent industrial documentation, spearheading the maintenance and inspection engineering services into the Industry 4.0 revolution. Just like the invention of the printing press, bringing digital tools and AI into industrial workflows will no doubt have far reaching impacts in order to build tomorrow’s future. 

 

This article is part of a series involving the participants of the AI program by Leonard, the VINCI Group’s foresight and innovation platform. The program has been specifically designed to accelerate the adoption of AI technologies within VINCI. It consists in a five-month incubation period where selected VINCI collaborators follow a learning by doing process where they codevelop an AI-based use case under the coaching and mentoring of Leonard’s team and consultants from eleven strategy.   

More about Leonard’s AI program: https://leonard.vinci.com/en/programmes/ia-course/ 

What do we do?

The VINCI Group created Leonard to tackle the challenges posed by the transformation of regions and lifestyles. Our goal is to unite a community of key stakeholders in order to build the city of the future together.