نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه مهندسی ماشینهای کشاورزی، دانشکده مهندسی و فناوری کشاورزی، دانشکدگان کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران.
2 گروه مهندسی ماشین های کشاورزی، دانشکده مهندسی و فناوری کشاورزی، دانشکدگان کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران
3 گروه مهندسی ماشین های کشاورزی، دانشکدۀ مهندسی و فناوری کشاورزی، دانشکدگان کشاورزی و منابع طبیعی، دانشگاه تهران، کرج، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Marine cage aquaculture along the southern coasts of Iran (Bushehr, Hormozgan, Khuzestan, and Sistan and Baluchestan) has experienced an annual growth rate of 15–20% in recent years and constitutes a key component of the national strategy to increase total aquaculture production to over 1.8 million tons. Hormozgan serves as the primary hub for marine species production; Bushehr produces approximately 14,000 tons with more than 20 active farms; Khuzestan targets 5,000 tons; and Sistan and Baluchestan benefits from the substantial potential of the Makran coast and the Oman Sea. Despite climatic constraints, infrastructural limitations, and market volatility, the sector’s high regional capacity and governmental support have positioned it as a strategic pillar of sustainable fisheries development in Iran. This study develops a scenario-based multi-objective mixed-integer linear programming (MILP) model to design the aquaculture supply chain under uncertainty. The model simultaneously addresses economic (minimization of total supply chain costs), social (maximization of sustainable employment), and environmental (reduction of fuel consumption and emissions) objectives. Uncertainty is incorporated through three scenarios—optimistic, moderate, and pessimistic. The model is implemented in Python using the PuLP library, and three solution approaches are compared: the ε-constraint method, the weighted-sum method, and robust optimization. Findings indicate that the weighted-sum approach provides the most practical and balanced solution, activating only the strategic node of Chabahar. This configuration yields a total supply chain cost of approximately 7,418,500 million IRR, generates 390,000 sustainable jobs, and results in energy consumption of 830,000 GJ. In contrast, the ε-constraint and robust optimization methods fail to produce feasible solutions due to rigid constraints and high sensitivity to pessimistic scenarios, respectively. Accordingly, the weighted-sum method is recommended as a flexible and context-appropriate approach for southern Iran. However, the relatively high energy consumption underscores the need to increase the environmental weight (w3=0.3) to enhance sustainability under uncertainty.
کلیدواژهها [English]
The primary objective of this study is to evaluate and compare various multi-objective optimization models for the design of a supply chain in marine cage aquaculture under conditions of uncertainty. The study focuses on achieving a balanced integration of three key objectives: economic efficiency (minimizing total costs), social sustainability (maximizing sustainable employment opportunities), and environmental responsibility (minimizing energy consumption and pollution levels). The overarching goal is to develop a resilient, efficient, and sustainable supply chain network that can adapt to the dynamic and uncertain conditions prevalent in the marine aquaculture industry. This is particularly relevant for the southern coastal regions of Iran, including the provinces of Hormozgan, Bushehr, Khuzestan, and Sistan and Baluchestan, which are characterized by a range of environmental, economic, and infrastructural challenges. These regions face significant uncertainties such as fluctuating market demands, climatic risks (e.g., temperature changes, storms, and sea level rise), limited production capacities, and inadequate infrastructure. By addressing these challenges, the study aims to support the formulation of long-term policies that promote sustainable development in the aquaculture sector.
The study emphasizes the importance of integrating multiple objectives into the supply chain design process. In traditional supply chain models, the focus is often on a single objective, such as cost minimization. However, in the context of marine cage aquaculture, it is essential to consider the interplay between economic, social, and environmental factors. For instance, while minimizing costs is crucial for the financial viability of the industry, it must be balanced against the need to create sustainable employment opportunities for local communities and to reduce the environmental footprint of aquaculture operations. Therefore, the study adopts a multi-objective approach to ensure that the supply chain design is not only economically efficient but also socially inclusive and environmentally responsible.
To achieve this, the research utilizes real-world data from both active and potential production nodes along the southern coasts of Iran. These nodes represent key locations where marine cage aquaculture can be implemented, and they vary in terms of geographical location, depth suitability, accessibility, and other relevant factors. By incorporating real data, the study ensures that the proposed models are grounded in practical realities and can be applied to actual industry conditions. This data-driven approach enhances the relevance and applicability of the findings, making them valuable for policymakers, industry stakeholders, and researchers.
The study also addresses the issue of uncertainty, which is a critical factor in the design of supply chains for marine aquaculture. Uncertainty can arise from various sources, including unpredictable changes in market demand, climate variability, and operational risks. To account for these uncertainties, the study employs a scenario-based approach, which allows for the evaluation of different possible future conditions. This approach enables the development of supply chain strategies that are robust and adaptable to a wide range of scenarios, thereby increasing the resilience of the aquaculture industry.
To achieve the objectives of the study, researchers developed and implemented a scenario-based multi-objective mixed-integer linear programming (MOMILP) model. This model is designed to incorporate the key uncertainties associated with marine cage aquaculture by considering three distinct scenarios: optimistic (increased demand), normal (stable conditions), and pessimistic (decreased demand with heightened risks). Each scenario represents a different possible future state of the industry, and the model evaluates the performance of the supply chain under each scenario to identify the most robust and effective solutions.
The MOMILP model was implemented in Python using the PuLP library, a powerful tool for solving linear programming problems. The model was used to evaluate and compare three different optimization approaches: the ε-constraint method, the weighted sum method, and robust optimization. Each of these methods has its own strengths and limitations, and the study aims to determine which method is most suitable for the specific context of marine cage aquaculture in Iran.
The ε-constraint method is a widely used approach in multi-objective optimization. In this method, one objective is selected as the primary objective to be optimized, while the other objectives are converted into constraints with specified bounds. This allows for the generation of a set of Pareto-optimal solutions, which represent the best possible trade-offs between the different objectives. However, the ε-constraint method can be computationally intensive, especially when dealing with a large number of constraints and variables.
The weighted sum method, on the other hand, is a more straightforward approach that aggregates the multiple objectives into a single objective function using predefined weights. In this study, the weights assigned to the economic, social, and environmental objectives were 0.5, 0.3, and 0.2, respectively. This method is computationally efficient and allows for the generation of a single optimal solution that reflects the priorities defined by the weights. However, the weighted sum method can be sensitive to the choice of weights, and it may not always produce a diverse set of solutions.
The third approach evaluated in the study is robust optimization, which is designed to find solutions that perform well across all scenarios. This method seeks to minimize the worst-case deviations from the optimal solution, ensuring that the supply chain remains effective even under the most adverse conditions. Robust optimization is particularly useful in situations where the level of uncertainty is high, as it provides a degree of protection against unexpected changes in the environment.
To ensure that the models are aligned with the local conditions of the southern coastal regions of Iran, the study analyzed data from 11 production nodes. These nodes were selected based on a range of factors, including their geographical location, depth suitability, accessibility, availability of subsidies, emission rates, and employment ratios. By considering these factors, the study ensures that the proposed supply chain strategies are tailored to the specific needs and constraints of the region. This localized approach enhances the practicality and feasibility of the solutions, making them more likely to be adopted by industry stakeholders.
The results of the study reveal that the weighted sum method was the most effective approach for the multi-objective optimization of the marine cage aquaculture supply chain. This method successfully generated an optimal solution that activated a single key production node, such as Chabahar, to achieve a production capacity of 100,000 tons. The solution demonstrated a well-balanced trade-off between the three objectives: economic efficiency (controlled total costs), social sustainability (high employment generation), and environmental responsibility (manageable energy consumption). The weighted sum method was able to produce this solution in under 0.02 seconds of computation time, which is a significant advantage in terms of efficiency and practicality.
In contrast, the ε-constraint method proved to be infeasible in this context. The method was unable to generate a viable Pareto front due to the stringent bounds placed on the minimum employment and maximum energy consumption. While the method estimated a high potential for employment generation, the constraints imposed by the model made it impossible to find a feasible solution. This highlights the limitations of the ε-constraint method in situations where the trade-offs between objectives are highly constrained.
Similarly, the robust optimization approach also resulted in infeasibility under conservative regret thresholds. The method was sensitive to the pessimistic scenarios, which led to a lack of feasible solutions. This sensitivity suggests that the model may require parameter adjustments, such as relaxing the bounds on certain constraints or reducing the level of conservatism, to accommodate the regional capacities and risks. These findings indicate that while robust optimization is a valuable approach in uncertain environments, it may not always be the most practical solution for the specific context of marine cage aquaculture in Iran.
The weighted sum method has been identified as the most effective and practical approach for designing a multi-objective supply chain in marine cage aquaculture under conditions of uncertainty. This method not only provides balanced solutions that consider economic, social, and environmental objectives but also aligns well with the specific conditions of Iran’s southern coastal regions. These areas face a unique set of challenges, including fluctuating market demands, environmental risks, and infrastructural limitations, which the weighted sum method effectively addresses by allowing for flexible prioritization through adjustable weights.
One of the key advantages of this method is its ability to adapt to different policy priorities by simply modifying the assigned weights to economic, social, and environmental goals. This flexibility enables decision-makers and stakeholders to tailor the supply chain design to their specific needs, whether it be maximizing employment, minimizing costs, or reducing environmental impact. As a result, the method supports the development of more resilient and adaptive aquaculture systems that can better withstand external shocks such as market volatility or climate-related disruptions.
Based on the findings, it is recommended that the weighted sum method be adopted as a standard tool for formulating sustainable aquaculture policies in Iran and other emerging economies facing similar challenges. Future improvements could involve integrating additional scenarios or hybrid optimization techniques to further enhance the model’s robustness and applicability in diverse and complex environments.
The study was funded by the University of Tehran, Country Iran and Project No. 323390.6.26.
Conceptualization, M.Sh. and H.S.Sh.; methodology, H.S.Sh. and M.Sh.; software, H.S.Sh.; validation, M.Sh., H.S.Sh. and A.A.; formal analysis, M.Sh., H.S.Sh.; investigation, H.S.Sh.; resources, H.S.Sh. and M.Sh.; data curation, H.S.Sh.; writing - original draft preparation, H.S.Sh., M.Sh. and M. Kh.; writing - review and editing, H.S.Sh., M.Sh., A.A. and M.Kh.; visualization, H.S.Sh.; supervision, M.Sh. and A.A.; project administration, M.Sh.; funding acquisition, M.Sh.
All authors have read and agreed to the published version of the manuscript.
Data available on request from the authors.
This research was conducted as part of the approved project No. 323390.6.26 by the Vice Presidency for Research and Technology, Faculty of Agricultural Engineering and Technology, College of Agriculture and Natural Resources, University of Tehran. We sincerely appreciate their financial and scholarly support. Additionally, we would like to express our appreciation to the Department of Agricultural Machinery Engineering, College of Agriculture and Natural Resources, University of Tehran, for providing us with essential technical and workshop facilities.
This study did not involve human or animal subjects, experimental procedures, or sensitive data; consequently, ethical approval was deemed unnecessary. The authors upheld the standards of academic integrity throughout the conduct and reporting of this research.
The authors declare no conflict of interest.
)