Case Study: Transformer-Based Solution for the Automatic Digitization of Gas Plants

TL;DR

The paper tackles automatic digitization of gas plant documentation, a problem worsened by heterogeneous P&ID formats and incomplete standardization. It proposes an end-to-end pipeline that combines OCR, Vision-Language models, a novel EGRTR scene-graph transformer, multi-expert fusion, and an optimization framework constrained by UNI 9167 regulations to produce both a structured design data view and a hierarchical plant graph. Key contributions include the EGRTR architecture, OCR-based matching with equipment lists, and a regression- and structure-aware post-processing algorithm that aligns outputs with regulatory rules and registry priors. Empirically, the system achieves high textual extraction accuracy (design data 91%, equipment matching 80%, specifications 84%) and solid scene-graph performance (e.g., mAP up to 0.762 at IoU 0.5 and component-level accuracy 95%), enabling practical plant digitization with room for improvement on complex diagrams. The workflow reduces manual effort for MGM users and demonstrates potential applicability to other technical documentation domains.

Abstract

The energy transition is a key theme of the last decades to determine a future of eco-sustainability, and an area of such importance cannot disregard digitization, innovation and the new technological tools available. This is the context in which the Generative Artificial Intelligence models described in this paper are positioned, developed by Engineering Ingegneria Informatica SpA in order to automate the plant structures acquisition of SNAM energy infrastructure, a leading gas transportation company in Italy and Europe. The digitization of a gas plant consists in registering all its relevant information through the interpretation of the related documentation. The aim of this work is therefore to design an effective solution based on Artificial Intelligence techniques to automate the extraction of the information necessary for the digitization of a plant, in order to streamline the daily work of MGM users. The solution received the P&ID of the plant as input, each one in pdf format, and uses OCR, Vision LLM, Object Detection, Relational Reasoning and optimization algorithms to return an output consisting of two sets of information: a structured overview of the relevant design data and the hierarchical framework of the plant. To achieve convincing results, we extend a state-of-the-art model for Scene Graph Generation introducing a brand new Transformer architecture with the aim of deepening the analysis of the complex relations between the plant's components. The synergistic use of the listed AI-based technologies allowed to overcome many obstacles arising from the high variety of data, due to the lack of standardization. An accuracy of 91\% has been achieved in the extraction of textual information relating to design data. Regarding the plants topology, 93\% of components are correctly identified and the hierarchical structure is extracted with an accuracy around 80\%.

Figures (5)

• Figure 1: Diagram of the developed solution. All the input data are in pdf format
• Figure 2: Examples of design data (a), page of equipment list (b) and P&ID (c)
• Figure 3: Developed pipeline for synthetic data generation
• Figure 4: EGRTR architecture
• Figure 5: (a) Example of inference by EGRTR for the recognition of components in a test P&ID. (b) Example of inference by EGRTR for the identification of relationships (in red) between recognised components in a test P&ID.