Requirement mining for model-based product design

Abstract

PLM software applications should enable engineers to develop and manage requirements throughout the product's lifecycle. However, PLM activities of the beginning-of-life and end-of-life of a product mainly deal with a fastidious document-based approach. Indeed, requirements are scattered in many different prescriptive documents (reports, specifications, standards, regulations, etc.) that make the feeding of a requirements management tool laborious. Our contribution is two-fold. First, we propose a natural language processing (NLP) pipeline to extract requirements from prescriptive documents. Second, we show how machine learning techniques can be used to develop a text classifier that will automatically classify requirements into disciplines. Both contributions support companies willing to feed a requirements management tool from prescriptive documents. The NLP experiment shows an average precision of 0.86 and an average recall of 0.95, whereas the SVM requirements classifier outperforms that of naive Bayes with a 76% accuracy rate.Keywords: requirements; unstructured; extraction; classification; natural language processing; NLP; supervised learning; machine learning. 306 R. Pinquié et al.Reference to this paper should be made as follows: Pinquié, R., Véron, R., Segonds, F. and Croué, N. (2016) Frédéric Segonds is an Assistant Professor at Arts et Metiers ParisTech and member of the Product Design and Innovation Laboratory (LCPI). His research interests focus on the early stages of design collaboration, optimisation and collaborative design. This area includes the integration of stakeholders' core competences into the early stages of design, and providing methodologies and tools to support early product design.Nicolas Croué is the Solutions and Consulting Director for the company Keonys and has responsibility for developing business and technical activities in a range of domains including simulation, systems engineering, search-based applications, business intelligence and manufacturing so as to develop solutions for the factory of the future. After graduating from the ESILV in Mechanical and Systems Engineering, he successfully worked for the Petroleum French Institute, SAMTECH and LMS and became an expert in systems engineering. This paper is a revised and expanded version of a paper entitled 'Natural language processing of requirements for model-based product design with ENOVIA/CATIA