Periodic Reporting for period 2 - XMANAI (Explainable Manufacturing Artificial Intelligence)
Reporting period: 2021-11-01 to 2023-04-30
Artificial Intelligence (AI) is paving its way into mainstream applications within several industries, including manufacturing, thanks to increased data volumes, advanced algorithms, and improvements in computing power and storage. The decisions and predictions that can be potentially derived from AI-enabled systems are becoming much more profound, and in many cases, critical to organizations success and profitability. However, despite the indisputable benefits that AI can bring in society and in any industrial activity, humans typically have little insight about AI itself and even less concerning the knowledge on how AI systems make any decisions or predictions due to the so-called “black-box” effect. The inner workings of machine learning and deep learning are not exactly transparent, and as algorithms become more complicated, fears of undetected bias, mistakes, and miscomprehensions in the decision-making process naturally grow among manufacturers and practically any stakeholder. If not addressed properly, this lack of trust might jeopardize the full potential of AI.
XMANAI approach focuses on explainable AI that aims at making AI models transparent and interpretable and AI processes humanly understandable and actionable at multiple layers. This would greatly increase human trust, and as the goal of AI is to support and optimize processes, people must feel empowered and know how the system works.
The main goal of XMANAI is to place the indisputable power of Explainable AI at the service of manufacturing and human progress carving out a “human-centric”, trustful approach that is respectful of European values and principles, and adopting the mentality that “our AI is only as good as we are”. XMANAI will therefore support the transformation of the manufacturing value chain with ‘glass box’ models that are explainable to a ‘human in the loop’ and produce value-based explanations for data scientists, data engineers and business experts.
XMANAI approach focuses on explainable AI that aims at making AI models transparent and interpretable and AI processes humanly understandable and actionable at multiple layers. This would greatly increase human trust, and as the goal of AI is to support and optimize processes, people must feel empowered and know how the system works.
The main goal of XMANAI is to place the indisputable power of Explainable AI at the service of manufacturing and human progress carving out a “human-centric”, trustful approach that is respectful of European values and principles, and adopting the mentality that “our AI is only as good as we are”. XMANAI will therefore support the transformation of the manufacturing value chain with ‘glass box’ models that are explainable to a ‘human in the loop’ and produce value-based explanations for data scientists, data engineers and business experts.
The first project period has been mainly focused on the preliminary activities needed for the XMANAI Platform development. In particular, the basic ingredients of the XMANAI concept in the form of user journeys, data and technical requirements and the preliminary features of the Minimum Viable Product (MVP) that will drive the next implementation steps of the project were defined (D1.2). Both technical requirements as well as business requirements gathered from the project Demonstrators (elaborated into D6.1) were mapped. Each demonstrator, following a specific methodology that leverages on Trial Handbooks, also provided the initial information related to its own data sources and data acquisition methods that will interact with the XMANAI platform and feed the AI models.
The activities also addressed the architectural design for the asset management layer for of the overall XMANAI Platform (D2.1) and the design of the XMANAI AI bundles, in particular the processes that will be supported, the design of the components that will be implemented to support these processes, the foreseen high-level user interfaces and the rationale underpinning the relevant design decisions (D3.1). Then, the design of the XMANAI reference architecture was provided in the form of architecture blueprints of the components and manufacturing apps that will drive the next implementation steps of the project (D5.1).
The activities also addressed the architectural design for the asset management layer for of the overall XMANAI Platform (D2.1) and the design of the XMANAI AI bundles, in particular the processes that will be supported, the design of the components that will be implemented to support these processes, the foreseen high-level user interfaces and the rationale underpinning the relevant design decisions (D3.1). Then, the design of the XMANAI reference architecture was provided in the form of architecture blueprints of the components and manufacturing apps that will drive the next implementation steps of the project (D5.1).
Today AI is considered as a valuable ally for humans, able to expand their capacities in the new digital era and improve their cognition. Moreover, explainable AI will play a pivotal role in making machine decisions and behaviours rational to humans, to drive them to the correct decisions, so that the optimal performance is reached.
XMANAI and its technological platform can clearly contribute to the research activities of Europe for advancing AI in manufacturing, focusing on the definition of novel explainable AI models and algorithms that will transform AI from a black-box intruder to a friendly ally and collaborator of business experts, fostering a collaboration mentality that will eventually pay off in better products and services, cost reductions, productivity growth and more wise resource utilisation.
XMANAI will mainly contribute to the following strategic impacts.
Products and services usable in a wide range of manufacturing processes leading to agile production processes and improved quality of products and processes.
Nowadays, manufacturers are pushed to embrace AI and machine learning techniques to support several processes in manufacturing systems, especially in those processes that required complex decision making processes and a vast amount of background knowledge and that have been purely performed by humans so far. Nevertheless, when it comes to such operations, where human rationale plays a very important, and in most cases the decisive role, it is very hard for AI models and similar algorithms to be trusted.
XMANAI aspires to respond to these shortcomings, by providing a trustworthy and reliable infrastructure that will allow manufacturers to better and more efficiently integrate AI, machine learning and data analytics in processes which are not considered appropriate for those at the moment. The main aim is to improve and give birth to new products and services, accelerate production cycles, improve quality of products and operations and eventually transform traditional production systems to data-driven manufacturing systems of high agility becoming part of a larger ecosystem where collaboration is key and new value chains can be developed, based on trustworthy data and intelligence sharing.
Humans working together with Artificial Intelligence systems in optimal complementarity
AI in manufacturing support human in taking more informed decisions by offering the necessary tools and methods to explore numerous possibilities and recommend actions. AI and machine learning can derive more objective results, and given the proper configuration, safeguard ethics, which might in specific cases be neglected by humans in their decisions, either purposely or even by mistake. XMANAI aims to put AI next to humans so that both sides can provide benefits to each other, and perform their tasks in complete complementarity without cultivating the perception in humans that they lose control or that they are threatened by the AI models.
XMANAI starts with the core objective to make AI models explainable and to provide business experts with graph machine learning analytics that are easier to understand and to work with. These tools can therefore become trusted companions to business experts and decision makers in their everyday tasks, explaining why certain results have been suggested, under which conditions, the inputs and assumptions that were considered, all this in an understandable way.
At the same time XMANAI offers the tools for humans to optimise existing models and algorithms, making them more applicable and interpretable. XMANAI concepts and components will be offered through the XMANAI Platform, that will strengthen humans and AI collaboration and contribute to a better productivity, quality and efficiency of the manufacturing system.
XMANAI and its technological platform can clearly contribute to the research activities of Europe for advancing AI in manufacturing, focusing on the definition of novel explainable AI models and algorithms that will transform AI from a black-box intruder to a friendly ally and collaborator of business experts, fostering a collaboration mentality that will eventually pay off in better products and services, cost reductions, productivity growth and more wise resource utilisation.
XMANAI will mainly contribute to the following strategic impacts.
Products and services usable in a wide range of manufacturing processes leading to agile production processes and improved quality of products and processes.
Nowadays, manufacturers are pushed to embrace AI and machine learning techniques to support several processes in manufacturing systems, especially in those processes that required complex decision making processes and a vast amount of background knowledge and that have been purely performed by humans so far. Nevertheless, when it comes to such operations, where human rationale plays a very important, and in most cases the decisive role, it is very hard for AI models and similar algorithms to be trusted.
XMANAI aspires to respond to these shortcomings, by providing a trustworthy and reliable infrastructure that will allow manufacturers to better and more efficiently integrate AI, machine learning and data analytics in processes which are not considered appropriate for those at the moment. The main aim is to improve and give birth to new products and services, accelerate production cycles, improve quality of products and operations and eventually transform traditional production systems to data-driven manufacturing systems of high agility becoming part of a larger ecosystem where collaboration is key and new value chains can be developed, based on trustworthy data and intelligence sharing.
Humans working together with Artificial Intelligence systems in optimal complementarity
AI in manufacturing support human in taking more informed decisions by offering the necessary tools and methods to explore numerous possibilities and recommend actions. AI and machine learning can derive more objective results, and given the proper configuration, safeguard ethics, which might in specific cases be neglected by humans in their decisions, either purposely or even by mistake. XMANAI aims to put AI next to humans so that both sides can provide benefits to each other, and perform their tasks in complete complementarity without cultivating the perception in humans that they lose control or that they are threatened by the AI models.
XMANAI starts with the core objective to make AI models explainable and to provide business experts with graph machine learning analytics that are easier to understand and to work with. These tools can therefore become trusted companions to business experts and decision makers in their everyday tasks, explaining why certain results have been suggested, under which conditions, the inputs and assumptions that were considered, all this in an understandable way.
At the same time XMANAI offers the tools for humans to optimise existing models and algorithms, making them more applicable and interpretable. XMANAI concepts and components will be offered through the XMANAI Platform, that will strengthen humans and AI collaboration and contribute to a better productivity, quality and efficiency of the manufacturing system.