Collaborative knowledge management to identify data analytics opportunities in additive manufacturing

Abstract

Additive Manufacturing (AM) is becoming data-intensive. The ability to identify Data Analytics (DA) opportunities for effective use of AM data becomes a critical factor in the success of AM. To successfully identify high-potential DA opportunities in AM requires a set of distinctive interdisciplinary knowledge. This paper proposes a methodology that enables collaborative knowledge management for identifying and prioritizing DA opportunities in AM. The framework of the proposed methodology has three components: a team of experts, a DA Opportunity Knowledge Base (DOKB), and a prioritization tool. The team of experts provides diverse knowledge that can be used to identify and prioritize DA opportunities. The DOKB, developed by using the Web Ontology Language (OWL), captures diverse knowledge from the experts to identify DA opportunities. The prioritization tool ranks the identified DA opportunities by using the Fuzzy integrated Technique of Order Preference Similarity to the Ideal Solution (Fuzzy-TOPSIS). A case study, in which National Institute of Standards and Technology (NIST) researchers participated, demonstrates our methodology. As a result, 264 DA opportunities for AM's Laser-Powder Bed Fusion (L-PBF) process are identified and prioritized. The prioritized DA opportunities help set a DA direction for L-PBF AM. Our methodology keeps knowledge sharable, reusable, revisable, and extendable. Thus, this methodology can continue to facilitate collaboration within the AM community to identify high potential and high impact DA opportunities in AM.

Development of a maturity model for additive manufacturing: A conceptual model proposal

“…In addition, additive manufacturing involves an increasing amount of data that is important for decision making [ 83 ]. Data Analytics can contribute to facilitate the decision-making process [ 84 ] and to the identification and prioritisation of opportunities [ 83 ] in additive manufacturing by extracting and generating a large amount of relevant information. However, this requires interdisciplinary knowledge and consequently the existence of different experts, whose coordination is not always easy [ 83 ].…”

“…Data Analytics can contribute to facilitate the decision-making process [ 84 ] and to the identification and prioritisation of opportunities [ 83 ] in additive manufacturing by extracting and generating a large amount of relevant information. However, this requires interdisciplinary knowledge and consequently the existence of different experts, whose coordination is not always easy [ 83 ]. Park et al developed a methodology to overcome this barrier and enable the management of different experts' knowledge, which takes into account the Collaborative Knowledge Management approach [ 83 ].…”

“…However, this requires interdisciplinary knowledge and consequently the existence of different experts, whose coordination is not always easy [ 83 ]. Park et al developed a methodology to overcome this barrier and enable the management of different experts' knowledge, which takes into account the Collaborative Knowledge Management approach [ 83 ]. This methodology incorporates three different components, a set of experts, the Data Analytics Opportunity Knowledge Base, and a prioritisation tool [ 83 ].…”

Knowledge-Based Data Identification for Machine Learning Use Cases

“…DfAM aims to design or redesign parts, products and components for additive manufacturing with 3D printers for more cost-effective, faster and efficient production (Gibson et al, 2021). At the same time, AM keeps becoming more data-intensive, generating an increasing amount of newly available data (Park et al, 2021). Given the availability of data and the benefits of AM, DfAM set out to properly exploit the potential of AM in product manufacturing.…”

Description Logic Ontology-Supported Part Orientation for Fused Deposition Modelling

“…During the past two decades, the application of DL ontologies in AM has gained importance and popularity. Many researchers developed ontologies or ontology-supported approaches to assist certain tasks in AM: Yim and Rosen [8] presented an ontology-supported case-based reasoning approach to assist AM process planning; Yim and Rosen [9] developed a ontology-based repository for AM design problems; Liu and Rosen [10] proposed an ontology-supported knowledge modelling and reuse approach for AM process planning; Witherell et al [11] constructed an ontology-based metamodel for composable and reusable laser powder bed fusion process; Eddy et al [12] developed an ontology-based intelligent tool for AM knowledge management; Roh et al [13] constructed an ontologybased laser and thermal metamodel for laser powder bed fusion; Lu et al [14] presented a set of ontology-supported digital solutions for integrated and collaborative AM; Assouroko et al [15] proposed an ontology-supported approach for characterising model fidelity in laser powder bed fusion; Dinar and Rosen [16] developed a design for AM ontology; Kim et al [17] proposed an ontology-based approach to link AM design to AM process planning; Hagedorn et al [18] presented an ontology-supported approach for innovative design for AM; Liang [19] proposed an ontology-oriented knowledge methodology for AM process planning; Kim et al [20] developed a design for AM ontology to support manufacturability analysis; Sanfilippo et al [21] constructed an ontology to represent the data and knowledge in the AM value chain; Ali et al [22] developed a product life cycle ontology for AM; Xiong et al [23] established an ontology-supported process planning framework for wire arc AM; Ko et al [24] studied machine learning and ontology based design rule construction for laser powder bed fusion; Chen et al [25] studied ontology-driven learning of Bayesian network for causal inference and quality assurance in laser powder bed fusion; Roh et al [26] established an ontology-based process map for laser powder bed fusion; Mayerhofer et al [27] studied ontology-driven manufacturability analysis for lithography-based ceramic manufacturing; Jarrar et al [28] presented an ontology-based approach for a decision support system in AM; Park et al [29] studied ontology-supported collaborative knowledge management to identify data analytics opportunities in laser powder bed fusion; Li et al…”