ModelWriter: Text & Model-Synchronized Document Engineering Platform
Ferhat Erata, Claire Gardent, Bikash Gyawali, Anastasia Shimorina, Yvan Lussaud, Bedir Tekinerdogan, Geylani Kardas, Anne Monceaux
TL;DR
This paper presents ModelWriter, a platform for text and model-synchronized document engineering that enables automated traceability analysis of technical documents, exemplified on Airbus SIDP. It combines semantic parsing to map natural language to Description Logic formulae, DL theorem proving to infer traces such as conflicts, contains, refines, equals, and requires, and relational-model finding to deduce additional trace links under configurable trace semantics in First-order Relational Logic. The authors demonstrate robust semantic parsing on 960 SIDPs (≈97.5% parsed) and high DL-axiom correctness (≈96%), and show practical inference performance with KodKod as the fastest backend among Alloy, KodKod, and Z3. The work advances industrial traceability by enabling new trace types, dynamic propagation of traces, and automatic detection of inconsistencies, with open-source tooling and applicability to safety-critical domains.
Abstract
The ModelWriter platform provides a generic framework for automated traceability analysis. In this paper, we demonstrate how this framework can be used to trace the consistency and completeness of technical documents that consist of a set of System Installation Design Principles used by Airbus to ensure the correctness of aircraft system installation. We show in particular, how the platform allows the integration of two types of reasoning: reasoning about the meaning of text using semantic parsing and description logic theorem proving; and reasoning about document structure using first-order relational logic and finite model finding for traceability analysis.
