<p>An effective evaluation method for analyzing the actual environmental multi-index systems is an important task. However, the inherent complexity and interrelatedness of environmental factors pose substantial challenges to this task. Many methods have been proposed and verified to solve this problem. However, unreasonable weight determination and sole single-index assessments limit the practical application of these analysis methods. In this study, we propose an optimally improved entropy weight calculation method (OIEW) that combines mathematical programming with the principle of consistency in entropy weight variation to determine the weight of each index. The simulation results demonstrate that our proposed method enhances robustness against extreme data while also effectively mitigating the over-correction of normal data during the weight determination process. Furthermore, by combining the OIEW method with the fuzzy comprehensive evaluation (FCE) method, the OIEW-FCE approach can be utilized to evaluate soil quality grades. Soil data from ten Chinese provinces are selected as the research specimens in this paper. An evaluation system for soil physical and chemical properties is developed, comprising two first-grade indices and six second-grade indices. The evaluation results show that the OIEW-FCE method significantly reduced the overall error in soil grade evaluation by approximately 20% compared to the improved entropy weight evaluation system utilizing the FCE and the OIEW methods employing the single factor evaluation method. This result indicates that our evaluation system can maintain accuracy and reliability in practical applications. Our method quantitatively assesses performance deviations from actual soil usage scenarios and has potential applications in relevant fields, such as environmental impact assessment, ecological resource management, and sustainable development planning.</p>