Hajiaghaeikeshteli, MostafaGholiyanjouybari, Fatemeh2024-12-172024-12-02Gholiyanjouybari, F. (2024). Designing sustainable agri-food supply chain networks [Tesis doctoral]. Instituto Tecnológico y de Estudios Superiores de Monterrey. Recuperado de: https://hdl.handle.net/11285/702932https://hdl.handle.net/11285/702932https://doi.org/10.60473/ritec.9https://orcid.org/0000-0002-9988-2626Agri-food products are critical to sustaining human life, as they provide essential nutrients for maintaining bodily functions. Agri-food production must match potential demand to ensure efficient supply to industries and markets. In terms of national and international concerns, this is one of the most important, even a top priority. In recent years, the agri-food industry has prioritized the development of efficient supply chain systems based on current trends and principles, such as sustainability and circular economy. The principles of sustainability can be employed to effectively utilize or reintroduce agri-food waste back into the network. This PhD dissertation deals with designing new agri-food supply chain networks for the first time in the literature. It not only considers the most important products to study, but also focuses on the recent trends and challenging issues like circular economy, water consumption, CO2 emission, and sustainability. We considered Saffron, Coconut, Soybean, and Pistachio, respectively, in four chapters of this thesis. In this work, we formulate some novel mixed-integer linear programming models to design agri-food supply chain networks in different agriculture industries, considering the above new challenges. The multi-objective networks struggle to manage the total net profit while monitoring CO2 emissions and the satisfaction of customers within the network. Given the NP-hard nature of the networks, the solution approach embraces a set of conventional, new, modified, and hybrid metaheuristics to surmount its complexity effectively. The effectiveness of the proposed mathematical models is certified by case studies and general problems evolved from real-world practices. In Saffron's work, we consider marketing practices and develop a stochastic multiobjective programming model to improve sustainability in three main areas. A convex robust optimization approach addresses farm production capacity uncertainty and saffron demand uncertainty. The LP-metric method is used to validate the mathematical model for the saffron business. We adopt a modified Keshtel algorithm to deal with the problem of NP-hardness. Two strategies are used to evaluate the performance of the proposed solution methods: a statistical comparison and a supportive tool that is based on multicriteria decision-making (MCDM). According to the MCDM method, MOKASEO outperformed other algorithms in small, medium, and large-sized problems compared to the other algorithms tested. The secodn supply chain network that we consider to design its colsed-loop network is for the coconut industry. We propose a new mixed-integer linear programming model to design an agri-food supply chain network under sustainable terms. With the goal of resolving a multi-objective closed-loop supply chain, both forward and reverse movements of products are taken into account. During the planning process, the model monitors environmental pollution within the network as well as job opportunities. Given the NP-hardness of the model, we use six multi-objective optimizers and three hybrid algorithms, among which the multi-objective artificial rabbit optimizer is first developed and applied in this study. Therefore, fifteen practical tests are conducted to determine whether the model is compatible with real conditions. The Friedman statistical test and interval plots demonstrate that optimizers are capable of solving problems of all sizes. In both statistical tests and the hybrid Multi-Criteria Decision Making (MCDM) framework, Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) outperformed solving practical tests. In the third work, we study circular economy in a closed-loop supply chain. To do this, we consider one of the most famous and valuable agricultural products, soybean. We formulate a novel mixed-integer linear programming model to design a closed-loop agri-food supply chain network under sustainability and circular economy terms. The multi-objective network strives to reduce CO2 emissions while monitoring customer satisfaction and overall net profit. Since the network is NP-hard, a combination of conventional and hybrid metaheuristics is used to overcome its complexity. Four multi-objective optimization algorithms and three hybrid algorithms are utilized to investigate the model's suitability for real-world conditions. A combination of interval plots and hybrid multi-criteria decision-making techniques demonstrates that optimizers can handle any size problem. For large and mediumsized problems, however, MOHHSA is more effective than MOGWO. Finally, in the fourth paper, we develop a new mixed linear mathematical model for the pistachio supply chain network to minimize the total fixed and variable costs of the closed-loop supply chain. This model is addressed with efficient and well-known meta-heuristic algorithms. A hybrid meta-heuristic algorithm is also developed to enhance the intensification and diversification phases. Finally, we compare and evaluate the quality of both meta-heuristic algorithms and hybrid algorithms.TextoengopenAccesshttp://creativecommons.org/licenses/by-nd/4.0INGENIERÍA Y TECNOLOGÍA::CIENCIAS TECNOLÓGICAS::TECNOLOGÍA DE LOS ALIMENTOS::NECESIDADES ALIMENTARIASTechnologyTesis Doctorado / doctoral ThesisFor at least three years because some of the ideas are being written to be published in the journlas.https://orcid.org/0000-0002-3358-9574Agri-food supply chainOptimizationSustainabilityCircular Economy114965