Matenga, AliceMurena, EriyetiMpofu, Khumbulani2025-03-042025-03-042020-01-012212-8271 (E)http://dx.doi.org/10.1016/j.procir.2020.03.101https://hdl.handle.net/20.500.14519/144330th CIRP Design 2020 (CIRP Design 2020).Digital transformation has ushered in the digital economy, powered by digital intelligence and quantum computing. The various winding topologies in rotary machines result from multi-variant design specifications and connection types. Rewinding of rotary machines is a behaviour-based decision making process conducted within the shop floor, as the procedure is dependent on multi-input multi-output variables. Due to high data variability in service remanufacturing of armature windings in rotary machines, data abstraction for intelligent automation and analytics leads to increased operational productivity and new insights into market dynamics. In this light, the aim of the paper is to illustrate the design of a prescriptive modelling system of a symmetrical multi-coil winding machine for armature winding. The proposed system is a hybrid least squares support vector machine and adaptive neuro fuzzy inference system for optimizing and maintaining a copper fill factor at 90.7%. A mixed method research was utilized for qualitative and quantitative for the multivariate parameters. The results show that the system through in-slot repetitive orthocyclic winding process, with multi-spindle concentric layering improves the energy efficiency of the induction motors, which in turn lowers winding faults during the remanufacturing process. Streamlining operations through fog computing further enhances system latency and process reliability towards sustainable industrialization.284-289 PagesenAttribution-NonCommercial-ShareAlike 4.0 Internationalhttp://creativecommons.org/licenses/by-nc-sa/4.0/Orthocyclic windingDeep learningSupport vector machineCollaborative designEnergy efficiencyPresentation