A self-adapted swarm architecture to handle big data for “factories of the future”

Currently, the manufacturing sector is facing a technological evolution with the so-called Industry 4.0. This poses a paradigm shift, enabling companies to be more competitive by taking advantage of innovative technologies(cloud computing, cyber-physical systems, big data analytics and deep learning), pursuing near-zero fault, near real-time reactivity to any problem, better traceability, more predictability in manufacturing, while working to achieve cheaper product customization. The challenges arise when the dimensionality of the data generated by manufacturing processes grows, affecting the performance of algorithms, decreasing it quickly as the dimension of the search space increases. Handling large datasets with a good performance in a limited time should be the main concern in Big Data analytics. This paper focuses on a logistic process of car manufacturing, where batteries are unloaded from trucks to warehouse, and then to the point of fit, where they are assembled into the car. It presents a complete data-driven architecture, using a swarm approach for distributed data processing among all data stages, where processing nodes with different tasks and technologies can work cooperatively to complete a job. The work presented in this paper is funded by the EU project BOOST4.0, focusing on a smart manufacturing scenario for the automotive sector.