Enhancing Machinery Design by Using Artificial Intelligence

Enhancing Machinery Design by Using Artificial Intelligence Juan Pablo Fiesco, Miguel Angel Mateo-Casali, Beatriz Andres(B), and Raul Poler Research Centre on Production Management and Engineering (CIGIP), Universitat Politècnica de València (UPV), Camino de Vera S/N, 46022 Valencia, Spain {jfiesco,mmateo,bandres,rpoler}@cigip.upv.es Abstract. This paper examines the significance of Industry 4.0 and artificial intelligence (AI) in the manufacturing sector, particularly by emphasising the role of design phase in the machinery life cycle. The design phase of a machine is a complex task that requires an advanced engineering and physics knowledge level. Nevertheless in the technology era, computer-aided design tools facilitate the design task. The area of data execution and simulation of machine behaviour in different scenarios is being researched and exploited by technologies, such as the Internet of Things (IoT) or AI. With this paper, three AI-based tools are proposed and conceptualised to support AI-assisted optimisation to generate design proposals to manufacture industrial equipment, structural components, mechanisms and control components. Keywords: artificial intelligence ·design life cycle ·equipment industry · Industry 4.0 1 Introduction In the modern Industry 4.0 environment, technological competences are crucial for companies to succeed in global markets. To optimise the added value in the design process of complex products, such as machinery with physical and dynamic elements, product development process management is being explored (Almoslehy & Alkahtani, 2021). The design function is responsible for creating a product that best meets customers’ needs, and it provides information on various aspects of the product before and after its production. Product design requirements encompass performance, reliability, size, cost, manufacturing, industry standards, government regulations, intellectual property and sustainability (Pfeifer, 2009). In the Industry 4.0 context, the generation and collection of large amounts of data in various formats throughout the product life cycle are becoming more commonplace. To make sense of this information, new data processing and analysis methods must be developed to transform data into easily understandable and explainable formats. This paper aims to identify what kind of needs manufacturing companies have during the process of designing industrial equipment and how these needs can be addressed with AI tools. Thus the following research questions arise: RQ1 What needs to exist in the industrial equipment design process? © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 J. Bautista-Valhondo et al. (Eds.): CIO 2023, LNDECT 206, pp. 342–347, 2024. https://doi.org/10.1007/978-3-031-57996-7_59 Enhancing Machinery Design by Using Artificial Intelligence 343 RQ2 What is the potential of using AI in the industrial equipment design? RQ3 What kind of AI tools can meet industrial equipment design needs? Accordingly, this paper is organised as follows: Sect. 2describes the followed methodology. Section 3examines the state of the art on needs in the design process by examining the link between AI during the product design process. Section 4proposes a set of AI tools to fulfil the identified needs. Finally, conclusions are discussed in Sect. 5. 2 Methodology The methodology used in this paper is based on reviewing the existing literature to identify the needs that appear in the design process by considering the current volatile characteristics of customer demand. The next step in the review focuses on how the digital technologies that support Industry 4.0 have been implemented into the design process to deal with short life cycle products. The paper is based on a conceptual case that describes the tools developed in the AIDEAS (2022) European Project, whose main aim is to develop AI technologies for supporting the entire life cycle of industrial equipment. This paper focuses on conceptualising AI tools to support the design phase of the life cycle in specific European machinery manufacturing companies. 3 State of the Art The design process involves continuous interplay between defining the problem and generating solutions, and multiple iterations required to achieve an optimal or robust design solution. The following subsections review design process needs and how these needs are fulfilled by applying industry 4.0 technologies. 3.1 Product Design Process Needs Enhancing the quality of decision making during product design and development is crucial, but can be challenging given the vast amount of available design information. Decision makers often struggle to access the right information they need at the right time, which makes the process time-sensitive (Riesener et al., 2019). The design process for mechanical products can be complex and time-consuming because of the numerous steps and improvement iterative activities that are involved. There is also pressure to reduce costs in a competitive market environment (Karayel et al., 2013). Traditional design approaches may need to be revised to adequately meet these needs. Today, product and system design not only satisfy user requirements and specifications, but also enable interactions between different components to initiate corrective actions whenever necessary by making operations and the entire product life cycle more intelligent (Hou et al., 2008). To adapt to the volatility of customer demands, customised product, shorter product life cycles and increased small batch production are emerging trends. 344 J. P. Fiesco et al. 3.2 Digitisation in Product Design Processes To meet volatile demands, new modern product design models emerge for promoting digital design and computer-aided technologies at every link of the product life cycle (Lei et al., 2022). To address the posed challenges, researchers have focused on establishing a cooperative and integrated environment using various computer-aided systems, including the Computer-Aided Design, Manufacturing and Engineering (CAD/CAM/CAE) application software, databases and web-based services (Saric et al., 2018). Although there are many support tools available for later design stages, such as detailed design, there are relatively very few tools available for initial conceptual stages (Meniru et al., 2003). Such lack of support can hinder a designer’s ability to manipulate, organise and represent design data by reducing their effectiveness. The development of Industry 4.0 has led to significant advances being made in digital twin (DT) technologies by opening the way to integrate AI, specifically data-driven machine-learning (ML) models. However, a recurring problem with these models is their susceptibility to train data and results lacking uniqueness (Farbiz et al., 2022). In the past, DT technology has been used primarily for tasks like fault detection, predicting maintenance needs and performance analysis. However, more attention should be paid to the potential of DT technology in product design. Specifically, exploring how the DT (virtual product) can improve the design process, and to lead to more efficient, informed and resourced results (Tao et al., 2019), has been limited. The DT technology can play a critical role in this process by providing a virtual representation of the physical product and facilitating the integration of Big Data and ML to support product design (Niu et al., 2021). As increasingly more design information is stored in databases, the integration of the DT with Big Data and AI can provide designers with a new level of understanding and rapid design decisions by means of AI algorithms to predict and adapt any kind of design proposals by providing designers with a competitive advantage in time-to-market terms. 4 AI to Support the Product Design Process The aim of this paper is to identify the appropriate instruments that can be used in the industrial machinery design stage within the scope of the European Project AI Driven Industrial Equipment Product Life Cycle Boosting Agility, Sustainability and Resilience (AIDEAS, 2022). The project seeks to enhance the sustainability, resilience and agility of European machinery manufacturing companies by leveraging AI-driven technologies to support the complete life cycle of industrial machinery, including its design, production, usage and maintenance. In the industrial equipment design phase, careful planning and execution are essential to ensure optimal performance and efficiency. By integrating AI technologies with CAD/CAM/CAE systems, designers can optimise the design of industrial equipment structural components, mechanisms and control components. Using these suitable tools can significantly enhance the key activities involved in the design phase, which can result in better machine performance and increased efficiency. Next a set of three tools is proposed and conceptualised to involve AI tools in the design phase of the machinery industry life cycle. Enhancing Machinery Design by Using Artificial Intelligence 345 The first proposed AI-based tool refers to as the Machine Design Optimiser (MDO). The MDO consists of an AI-powered tool that assists designers to define the key design parameters in multiphysical systems. AI is used to optimise design parameters of machines by employing techniques like evolutionary algorithms or metaheuristic algorithms to find optimal configurations that meet specified objectives, such as minimising energy use, maximising throughput or reducing material usage. AI in the MDO will enable algorithms to be designed that learn from existing designs and user-defined constraints to generate new design alternatives. AI-enabled simulations and virtual prototyping tools can be employed to simulate and evaluate different machine design variants. Finally, AI algorithms can analyse design specifications, perform simulations and compare designs to predefined criteria to detect design flaws or risks early in the development process. The second proposed AI-based tool focuses on data synthesis for training the MDO AI tool, and is called the Machine Synthetic Data Generator (MDG). The MDG synthesises high-quality datasets by simulations, which are essential for analysing machine design and training optimisation algorithms to propose optimal design parameters. As a result, AI solutions will soon be accessible for small-scale and short-term projects, and will require fewer resources to train the necessary algorithms. The MDG toolkit will enable to synthesise large high-quality datasets by simulations for the machine design analysis and for training the optimisation algorithms that will propose optimal design parameters. AI MDG solutions will be accessible for shorter time series and smaller volume productions, and will require fewer resources for training the necessary AI algorithms. To integrate AI-assisted optimisation modules with existing CAx systems, an interoperability system is essential. CAx Addon aims to develop AI-assisted optimisation modules (MDO and MDG) into products that can be easily integrated with standard CAD/CAM/CAE systems. This involves creating user interfaces and APIs that will allow seamless integrations with systems like SolidWorks by making them ready for use in real-world scenarios. APIs and UIs will incorporate the individual functionalities of the optimisation modules, but will also consider the requirements of other standard CAx solutions. These modules can significantly enhance machine performance and efficiency by improving industrial equipment designs. By employing AI-powered tools like the MDO and the MDG, designers can improve the industrial equipment design phase and optimise performance. 5 Conclusion The integration of AI tools into the equipment design phase offers numerous benefits, including the reduction of the design cycle time and waste of mass, and enhancing sustainability throughout the extended life cycle of industrial equipment. The paper conceptualises a set of AI-based toolkits, such as the MDO, the MDG and CAx Addon, which can further optimise design process by allowing the critical aspects of equipment to be optimally designed. Using AI tools intends to streamline the equipment design process and to improve the design features for a longer life cycle by allowing equipment to be put to good use. This initiative explains a first approach that defines the essential 346 J. P. Fiesco et al. characteristics of several AI tools that are employed to improve the machine life cycle with AI techniques. A future research line proposes identifying which type of AI techniques are the most suitable ones for developing these tools. Another research line is to identify and define the requirements of each tool at the user level for them to be useful. In line with this, this paper aims to be a basis of how AI can be applied in the machinery design process to, thus, be able to make a methodology or conceptual framework of different tools. Acknowledgements. The research that led to these findings received funding from two sources. The first source of funding was from the Horizon Europe Framework Programme (HORIZON) with Grant Agreement No. 101057294 “AI Driven Industrial Equipment Product Life Cycle Boosting Agility, Sustainability, and Resilience (AIDEAS)”. The second source of funding was from the Regional Department of Innovation, Universities, Science, and Digital Society of the Generalitat Valenciana “Programa Investigo” (ref. INVEST/2022/330), which the European Union supported - NextGenerationEU under the Plan de Recuperación, Transformación y Resiliencia. References AIDEAS (2022) AI Driven Industrial Equipment Product Life Cycle Boosting Agility, Sustainability and Resilience. European Union’s Horizon Europe Research and Innovation Programme under grant agreement No. 101057294 Almoslehy, S.A.M., Alkahtani, M.S.: Key approaches, risks, and product performance in managing the development process of complex products sustainably. Sustain. (Switz.) 13(9), 4727 (2021). https://doi.org/10.3390/su13094727 Farbiz, F., Habibullah, M.S., Hamadicharef, B., Maszczyk, T., Aggarwal, S.: Knowledgeembedded machine learning and its applications in smart manufacturing. J. Intell. Manuf. (2022). https://doi.org/10.1007/s10845-022-01973-6 Hou, J., Su, C., Wang, W.: Knowledge management in collaborative design. In: Proceedings of 2008 IEEE International Conference on Service Operations and Logistics, and Informatics, IEEE/SOLI 2008, vol. 1, pp. 848–852 (2008). https://doi.org/10.1109/SOLI.2008.4686517 Karayel, D., Ozkan, S.S., Vatansever, F.: Integrated knowledge-based system for machine design. Adv. Mech. Eng. 2013, 1–10 (2013). https://doi.org/10.1155/2013/702590 Lei, Y., Vyas, S., Gupta, S., Shabaz, M.: AI based study on product development and process design. Int. J. Syst. Assur. Eng. Manage. 13, 305–311 (2022). https://doi.org/10.1007/s13198021-01404-4 Meniru, K., Rivard, H., Bédard, C.: Specifications for computer-aided conceptual building design. Des. Stud. 24(1), 51–71 (2003). https://doi.org/10.1016/S0142-694X(02)00009-1 Niu, X., Wang, M., Qin, S.: Product design lifecycle information model (PDLIM). Int. J. Adv. Manuf. Technol. 118, 2311–2337 (2021). https://doi.org/10.1007/s00170-021-07945-z/Pub lished Pfeifer, M.: Design requirements. In: Materials Enabled Designs, pp. 23–50. Elsevier (2009). https://doi.org/10.1016/b978-0-7506-8287-9.00002-1 Riesener, M., Schuh, G., Dölle, C., Tönnes, C.: The digital shadow as enabler for data analytics in product life cycle management. Procedia CIRP 80, 729–734 (2019). https://doi.org/10.1016/j. procir.2019.01.083 Enhancing Machinery Design by Using Artificial Intelligence 347 Saric, I., Pervan, N., Muminovic, A., Colic, M.: Development of integrated intelligent cad system for design of shafts. Tehnicki Vjesnik 25, 99–104 (2018). https://doi.org/10.17559/TV-201705 21194820 Tao, F., et al.: Digital twin-driven product design framework. Int. J. Prod. Res. 57(12), 3935–3953 (2019). https://doi.org/10.1080/00207543.2018.1443229