Integrated Rough AHP and Neural Network Model for Mobile Phone Selection with Big Data Under Uncertainty

Abstract

This paper applied an integrated approach to Multi-Attribute Decision Making (MADM) by combining the Rough Analytic Hierarchy Process (RAHP) and Neural Network, specifically a Multi-Layer Perceptron (MLP) for a specific problem of smartphone selection. The Rough Analytic Hierarchy Process, grounded in rough set theory, proves adept at handling uncertainties in decision-making processes. Through the integration of RAHP and MLP, this study provides a comprehensive framework for ranking mobile phone criteria, focusing on camera quality, selfie capabilities, audio performance, display features, battery life, and pricing. The practical example employed demonstrates the applicability of the proposed methodology in real-world decision-making scenarios, the fusion of RAHP and MLP emerges as a potent solution for Multiple Attribute Decision Making (MADM) problems, offering decision-makers confidence in navigating intricate scenarios. This integrated approach signifies a new era of robust decision-making, enhancing outcomes across diverse domains by synergizing structured prioritization and uncertainty management. The paper proceeds with a literature review, outlining existing approaches in decision-making scenarios. The methods section details the operations with rough numbers, the Rough Analytic Hierarchy Process, and the Multi-Layer Perceptron. A numerical example of mobile phone selection is presented, illustrating the application of the integrated approach. In the presented numerical example, two scenarios are provided: one without a price criterion and another with a price criterion. In the price-less scenario, the Honor Magic5 Pro is chosen, while in the scenario considering price, the Oppo Find X6 Pro is selected as the best option.