Stereoscopic scalable quantum convolutional neural network-based personalized ranking of sustainable suppliers’ selection

This paper proposes a comprehensive framework for achieving sustainable supplier selection by integrating advanced methods and models. First, correlation coefficients with min–max normalization were used as the preliminary phase to refine the data quality and pertinence. This normalization technique helped to standardize all analysis data sets to the same scale, which makes subsequent analysis more accurate. After preprocessing, the Hybrid Kookaburra Artificial Hummingbird Algorithm (HKAHA) was applied for feature selection to demonstrate the suitability of this algorithm in selecting the most essential criteria for supplier evaluation. Concerning the second research question, the HKAHA effectively coordinates the exploration and exploitation processes, resulting in a better selection of features regarding sustainable supplier performance. To augment decision-making, a Stereoscopic Scalable Quantum Convolutional Neural Network (SSQ-CNN) was proposed for supplier classification and supplier ranking. This novel structure of a neural network incorporates elements of quantum computing to maximize quantitative performance, allowing for much finer gradations in judging prospective suppliers based on their sustainability profile. Lastly, the Osprey Optimization Algorithm (OOA) was applied to adjust and improve the performance of the SSQ-CNN in the final supplier selection decision, thereby helping to produce sound decisions. This research study demonstrates that the proposed framework not only enhances the reliability of the suppliers’ ranking but also aligns with the corresponding sustainable development goals by focusing on suppliers who conform to economic, social, and environmental standards. The proposed approach achieves a higher accuracy of 99.95%.