Artificial intelligence models are increasingly used in manufacturing to inform decision making. Responsible decision making requires accurate forecasts and an understanding of the models’ behavior. Furthermore, the insights into the models’ rationale can be enriched with domain knowledge. This research builds explanations considering feature rankings for a particular forecast, enriching them with media news entries, datasets’ metadata, and entries from the Google knowledge graph. We compare two approaches (embeddings-based and semantic-based) on a real-world use case regarding demand forecasting. The embeddings-based approach measures the similarity between relevant concepts and retrieved media news entries and datasets’ metadata based on the word movers’ distance between embeddings. The semantic-based approach recourses to wikification and measures the Jaccard distance instead. The semantic-based approach leads to more diverse entries when displaying media events and more precise and diverse results regarding recommended datasets. We conclude that the explanations provided can be further improved with information regarding the purpose of potential actions that can be taken to influence demand and to provide “what-if” analysis capabilities.

Enriching Artificial Intelligence Explanations with Knowledge Fragments

Evaluating eXplainable artificial intelligence tools for hard disk drive predictive maintenance

This study addresses the emerging concept of Industry 5.0, which aims to tackle societal concerns associated with the ongoing digital industrial transformation. However, there is still a lack of consensus on the definition and scope of Industry 5.0, as well as limited understanding of its technological components, design principles, and intended values. To bridge these knowledge gaps, the study conducts a content-centric review of relevant literature and synthesizes evidence to develop an architectural design for Industry 5.0. The findings reveal that Industry 5.0 represents the future of industrial transformation, offering potential solutions to socio-economic and environmental issues that were inadequately addressed or exacerbated by Industry 4.0. The study provides managers, industrialists, and policymakers with a comprehensive overview of Industry 5.0, including its technological constituents, design principles, and smart components, emphasizing the importance of stakeholder involvement and integration for effective governance of digital industrial transformation within this framework. 

Behind the definition of Industry 5.0: a systematic review of technologies, principles, components, and values

Amplifying Employability Skills to Create Co-Working Space for Human and Cobots in the E-Commerce Industry

Quality control is a crucial activity performed by manufacturing enter-prises to ensure that their products meet quality standards and avoid potential damage to the brand’s reputation. The decreased cost of sensors and connectivity enabled increasing digitalization of manufacturing. In addition, artiﬁcial intelligence enables higher degrees of automation, reducing overall costs and time required for defect inspection. This research compares three active learning approaches (with single and multiple oracles) to visual inspection. We propose a novel approach to probabilities calibration of classiﬁcation models and two new metrics to assess the performance of the calibration without the need for ground truth. We performed experiments on real-world data provided by Philips Consumer Lifestyle BV . Our results show that explored active learning settings

Active Learning and Approximate Model Calibration for Automated Visual Inspection in Manufacturing

Human-centricity is the core value behind the evolution of manufacturing towards Industry 5.0. Nevertheless, there is a lack of architecture that considers safety, trustworthiness, and human-centricity at its core. Therefore, we propose an architecture that integrates Artificial Intelligence (Active Learning, Forecasting, Explainable Artificial Intelligence), simulated reality, decision-making, and users' feedback, focusing on synergies between humans and machines. Furthermore, we align the proposed architecture with the Big Data Value Association Reference Architecture Model. Finally, we validate it on three use cases from real-world case studies.

Human-Centric Artificial Intelligence Architecture for Industry 5.0 Applications

Exploiting virtual reality and the robot operating system to remote-control a humanoid robot

Natural Language Processing (NLP) is crucial to perform recommendations of items that can be only described by natural language. However, NLP usage within recommendation modules is difficult and usually requires a relevant initial effort, thus limiting its widespread adoption. To overcome this limitation, we introduce FORESEE, a novel architecture that can be instantiated with NLP and Machine Learning (ML) modules to perform recommendations of items that are described by natural language features. Furthermore, we describe an instantiation of such architecture to provide a service for the job market where applicants can verify whether their curriculum vitae (CV) is eligible for a given job position, can receive suggestions about which skills and abilities they should obtain, and finally, can obtain recommendations about online resources which might strengthen their CVs.

Rubén Alonso

Papers

Human-Centric Artificial Intelligence Architecture for Industry 5.0 Applications

Exploiting virtual reality and the robot operating system to remote-control a humanoid robot

A General and NLP-based Architecture to perform Recommendation: A Use Case for Online Job Search and Skills Acquisition