| تعداد نشریات | 282 |
| تعداد نشریات سازمان | 80 |
| تعداد شماره‌ها | 3,252 |
| تعداد مقالات | 35,930 |
| تعداد مشاهده مقاله | 54,702,401 |
| تعداد دریافت فایل اصل مقاله | 93,758,745 |
بهینه‌سازی وزن‌های تصمیم‌گیری چندمعیاره با الگوریتم ژنتیک برای انتخاب بهترین تیمار مدیریتی کلزا تحت مدیریت‌های مختلف آبیاری و محرک‌های رشد | ||
| پژوهش های خاک | ||
| مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 15 مهر 1405 | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22092/ijsr.2026.373973.833 | ||
| نویسندگان | ||
| رضا محمدی کیا* 1؛ محمد پسندیده2 | ||
| 1محقق، بخش مدیریت آب در مزرعه، مؤسسه تحقیقات خاک و آب، سازمان تحقیقات، آموزش و ترویج کشاورزی، کرج، ایران | ||
| 2استادیار، بخش شیمی، حاصلخیزی خاک و تغذیه گیاهی، مؤسسه تحقیقات خاک و آب، سازمان تحقیقات، آموزش و ترویج کشاورزی، کرج، ایران | ||
| چکیده | ||
| انتخاب بهترین تیمار مدیریتی در کشت کلزا به‌ویژه در شرایط تنش آبی مسئله‌ای چندبعدی است که نیازمند ارزیابی هم‌زمان معیارهای متعارضی مانند عملکرد دانه، بهره‌وری آب، درصد روغن و آب مصرفی می‌باشد. در مطالعه پیشین نگارندگان، تیمار اسید آمینه تحت تنش آبی (WS × Amino acid) با استفاده از روش ویکور (VIKOR) و وزن‌دهی یکسان به‌عنوان بهترین گزینه معرفی شده بود، اما پایداری این نتیجه در برابر تغییر روش تحلیل بررسی نشده بود. در این پژوهش، با هدف اعتبارسنجی و تکمیل آن یافته، هشت ترکیب تصمیم‌گیری حاصل از دو الگوریتم تاپسیس (TOPSIS) و ویکور (VIKOR) و چهار روش وزن‌دهی یکسان، آنتروپی شانون، کریتیک و الگوریتم ژنتیک به‌کار گرفته شد. نتایج نشان داد که هر هشت ترکیب بدون استثنا تیمار اسید آمینه تحت تنش آبی را به‌عنوان بهترین گزینه معرفی کردند و ضریب همبستگی اسپیرمن با میانگین 9/0 توافق بسیار بالا بین روش‌ها را تأیید نمود. تحلیل جداگانه شرایط آبیاری آشکار ساخت که اسید آمینه در شرایط تنش آبی و هیومیک اسید در شرایط آبیاری نرمال بهترین عملکرد را دارند. از منظر روش‌شناختی، ترکیب الگوریتم ژنتیک با ویکور با مقدار جداسازی 222/0 بالاترین قدرت تفکیک را ایجاد کرد که حدود پنج برابر بیشتر از ویکور با وزن یکسان بود. یافته‌های این پژوهش نشان داد که برتری تیمار اسید آمینه تحت تنش آبی پدیده‌ای مستقل از روش تحلیل است و بهینه‌سازی وزن‌ها با الگوریتم ژنتیک می‌تواند اطمینان تصمیم‌گیری را به‌طور قابل‌توجهی افزایش دهد. | ||
| کلیدواژه‌ها | ||
| بهره‌وری آب؛ تنش آبی؛ عملکرد دانه؛ کشاورزی پایدار؛ مدیریت آبیاری | ||
| عنوان مقاله [English] | ||
| Optimization of Multi-Criteria Decision-Making Weights Using a Genetic Algorithm for Selecting the Best Management Treatment in Canola under Different Irrigation Management Practices and Growth Promoters | ||
| نویسندگان [English] | ||
| Reza Mohammadikia1؛ mohammad Pasandideh2 | ||
| 1Researcher, On-Farm Water Management Department, Soil and Water Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj, Iran | ||
| 2Assistant Professor, Soil Chemistry, Fertility and Plant Nutrition Department, Soil and Water Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Karaj, Iran | ||
| چکیده [English] | ||
| Extended Abstract Introduction Canola (Brassica napus L.) is one of the most important oilseed crops globally, but its production in arid and semi-arid regions such as Iran is severely affected by water stress, leading to significant reductions in seed yield and oil content. Biostimulants such as amino acids, humic acid, fulvic acid, and seaweed extract have emerged as effective tools to mitigate the negative effects of environmental stresses. However, their effectiveness depends strongly on environmental conditions. Selecting the best "irrigation × biostimulant" combination is a multi-criteria decision-making problem involving conflicting criteria such as seed yield, water productivity, oil percentage, and water use. In the authors' previous study, the WS × Amino acid treatment was identified as the best option using VIKOR with equal weighting. However, the robustness of this result against changes in weighting methods and ranking algorithms had not been examined—leaving open the question of whether this superiority was method-dependent or inherent in the data. Materials and Methods Data for this study were derived from a field experiment in which two irrigation levels (normal and water stress) and five biostimulant levels (control, amino acid, humic acid, fulvic acid, and seaweed extract) were examined in a total of 10 treatments. Four evaluation criteria water productivity (kg m⁻³), seed oil percentage (%), water use (m³ ha⁻¹), and seed yield (kg ha⁻¹) were included in the decision matrix. For comprehensive evaluation and validation of the previous finding, eight decision-making combinations were employed, formed by crossing two ranking algorithms (TOPSIS and VIKOR) with four weighting methods (equal weighting, Shannon's entropy, CRITIC, and genetic algorithm). TOPSIS operates based on Euclidean distance from the ideal solution, while VIKOR is based on compromise and consensus logic. Genetic algorithm weights were optimized using NSGA-II to maximize the separation between the best and second-best alternatives. Robustness analysis was conducted from four perspectives: agreement on the best alternative, Spearman correlation for the overall ranking structure, separate ranking of stress and normal treatments, and independence from weighting philosophy and ranking algorithm. Results and Discussion The results showed that all eight decision-making combinations, without exception, identified the WS × Amino acid treatment as the best option a 100% agreement that is rare in the multi-criteria decision-making literature. The Spearman correlation coefficient with a mean of 0.904 confirmed very high agreement among the methods; indeed, six of the eight methods produced completely identical rankings (ρ = 1.0). Separate analysis of irrigation conditions revealed a clear pattern. Under water stress conditions, amino acid was the best stimulant, increasing seed yield by 23% (from 3002 to 3703 kg ha⁻¹) and water productivity by 28% (from 1.40 to 1.79 kg m⁻³) compared to the control. It was followed by seaweed extract (+18%), humic acid (+14%), and fulvic acid (+14%). Under normal irrigation conditions, however, humic acid was superior, increasing seed yield by 14% (from 3369 to 3844 kg ha⁻¹) and achieving the highest oil content (43.9%), while amino acid fell to fourth place. This condition-dependent ranking shift is one of the most important agronomic findings of this study. From a methodological perspective, GA-VIKOR achieved the highest discrimination power with a separation value of 0.222 approximately five times greater than VIKOR with equal weighting (0.044) and about six times greater than VIKOR with CRITIC weighting (0.038). GA-TOPSIS was the second-best method with 0.135. These results demonstrate that optimizing weights with a genetic algorithm, while preserving the final outcome, substantially increases decision-making confidence. Conclusion This study produced three main achievements: First, it proved the complete robustness of the previous study's finding, demonstrating that the superiority of WS × Amino acid is an inherent, method-independent phenomenon not the result of choosing a specific analytical approach. Second, it provided conditional management recommendations: under water-limited conditions, amino acid is the best choice, while under full water availability, humic acid should be preferred. Third, it demonstrated the superiority of GA-VIKOR in enhancing discrimination power and decision-making confidence, thereby promoting the application of integrated GA-MCDM approaches in agricultural sciences. These findings have direct practical implications for farmers, water managers, and agricultural policymakers. The proposed framework can be extended to other crops, other climatic regions, and combined with multi-objective algorithms for future research. | ||
| کلیدواژه‌ها [English] | ||
| Grain Yield, Irrigation Management, Sustainable Agriculture, Water Productivity, Water Stress | ||
|
آمار تعداد مشاهده مقاله: 1 |
||