Abstract: Rapid digital transformations in every aspect of life, work, and society are aided by quickly developing technologies like artificial intelligence (AI) and the Internet of Things (IoT). The sophisticated data-driven decision-making environment that they inspire enables the implementation of smart manufacturing, in which every aspect of manufacturing is monitored using sensors that continuously record and analyze data streams in real-time. The focus is on monitoring the real-time state of affairs within the ecosystem of machines, processes, and resources. This shift in paradigm can increase productivity, efficiency, and profitability in a significantly disruptive manner by enhancing the transparency of operations and automating data-driven decision-making processes. This transformation needs to embrace many technological and systematic alterations that require aligned collective efforts from stakeholders. The design, development, and interoperation are essential for the successful implementation of smart manufacturing systems. Mass customization, lower energy consumption, retrofitting and reusability of assets, lower environmental impact, and a more sustainable production process are desirable manufacturing efficiencies that will drive up the acceptance of smart manufacturing systems.
A supply chain (SC) must be designed to allow for efficient management of all aspects of supply chain planning, analysis, modeling, monitoring, and control with the support of data-driven business intelligence (BI) systems. The smart manufacturing system architecture that is detailed in this context. Singularity of the suggested system architecture shields processes, operations, systems, subsystems, and their interactions from external environment factors that have an effect on them, to provide a high-quality working mode in the time of overflowing demand. The core of the AI-assisted BI system centered around prediction includes (i) a HW/SW architectural setup, and (ii) and AI algorithms with differing depths for data cleansing and feature engineering that enables the existence of such a smart manufacturing system architecture. Alternative AI algorithms are employed with fusing/conjunction of numerous learning algorithms for more efficient training of models with superior forecast accuracy to predict production, delivery, and external demand. The AI-assisted BI system is scalable and adaptable to more than 2 massive datasets for expectant production planning and control through training of models that can provide estimates of machine UT, jobs carried out in time, and the number of finished products. Contemplating on the trade-off between profitability and sustainability with model operationalization that considers data governance, data utilization, and data development costs alongside carbon- and energy-aware manufacturing are possible.
Keywords: Sustainable Manufacturing, Artificial Intelligence, Supply Chain Optimization, Green Manufacturing, Predictive Analytics, Smart Logistics, Resource Efficiency, Carbon Footprint Reduction, AI-Driven Decision Making, Circular Economy, Environmental Impact, Machine Learning, Real-Time Monitoring, Energy Efficiency, Data-Driven Manufacturing.
Copyright Form