Optimization model for enterprise financial management utilizing genetic algorithms and fuzzy logic

PeerJ Computer Science, Год журнала: 2025, Номер 11, С. e2812 - e2812

Опубликована: Апрель 7, 2025

This study explores the complexities of enterprise financial management by optimizing models with a particular focus on enhancing risk prediction performance. A multi-objective mathematical model is first developed to establish key optimization goals, including cost reduction, improved capital utilization, and increased economic benefits. systematically defines decision variables objectives, providing comprehensive framework for management. To improve predictive accuracy, integrates genetic algorithms back-propagation (BP) neural networks, leveraging optimize network’s parameters structure. Additionally, hierarchical reinforcement learning based fuzzy reasoning (HRL-FR) proposed enhance decision-making capabilities. employs policy optimization, incorporating address uncertainties in complex dynamic environments. Experimental validation using Compustat dataset confirms effectiveness model. Key variables, working asset ratio debt-to-equity ratio, are identified as significant influencers reinforcing model’s robustness. The algorithm’s search process identifies parameter combinations that maximize network performance, further improving Comprehensive evaluations conducted Center Research Security Prices (CRSP) datasets 2022 confirm HRL-FR superior ability predict analyze information accurately. demonstrates higher profitability, enhanced efficiency, curves closely align optimal models. These findings highlight potential powerful tool offering valuable insights mitigation strategic decision-making.

Язык: Английский

New perspectives on university quality assessment: A Mamdani Fuzzy Inference System approach

PLoS ONE, Год журнала: 2025, Номер 20(5), С. e0321013 - e0321013

Опубликована: Май 19, 2025

Higher education has traditionally played the role of an overarching factor in economic growth and development. The implementation European Education Area (EHEA) already achieved improvements many educational areas, but there remain, within requirement to ensure academic excellence, cases where quality criteria are not entirely harmonized. Genuine harmonization among 48 countries that have so far been affiliated with EHEA a key challenge for national assessment agencies related bodies. This study aims analyze Spanish university system partially through model based on Mamdani Fuzzy Inference System (FIS) methodology. Numerous studies identified evaluate from perspective student, no public higher institutions faculty employees. research gap prompted extensive literature review, considering fifteen main elements classified into five categories: internationalization; scientific production, occupational category, background, professional experience. Researchers collected curated data database four Madrid-based institutions. A FIS, yielding unique each case, was implemented using MATLAB Logic Toolbox. Therefore, results evaluated determine which institution led better quality. approach leads measuring First, thanks evaluation workers professors who part universities Madrid. Second, we carried out this analysis under methodology used before issue. Concerning its practical implications, can help policymakers design practices improve careers and, as result, future employability graduates.

Язык: Английский

Hardware Design Space Exploration in High-Level Synthesis Backend Featuring Online Arithmetic

Опубликована: Сен. 16, 2024

Язык: Английский