Utilizing statistical significance in fuzzy interval-valued evidence sets for assessing artificial reef structure impact
基于墨西哥湾观测数据，用卡方检验和模糊区间值证据模型，比较不同人工鱼礁材料对鱼类存在和丰度的影响，发现油气平台结构影响最大，并考虑成本因素。
Artificial reefs are often formed by reef balls, oil rigs, either toppled or partially removed, and sunken vessels. Data gathered from observations made in the Gulf of Mexico off the Texas coast were statistically analyzed to investigate the impact of type of structure on fish presence, abundance and species observed. Environmental variables were controlled, and crosstabulations between various structure material-type categories were conducted for the most frequently observed species. Based on Chi-square tests of significance, a fuzzy interval-valued evidence model was developed such that different types of artificial reef structures were compared to determine to what extent each structure was likely to impact fish presence and abundance of species. Results indicate that all materials have been at least somewhat impactful, but the most impact on the presence of individual species has been the oil and gas jackets. A fuzzy goal-driven extension found that potential costs associated with each material may affect reef material selection decision making.