Supply chain network design in the agricultural food sector: Mathematical modelling based on the example of Taşköprü garlic

Unlike other sectors, the agri-food sector poses distinct risks arising from both agricultural production and food supply processes. Factors such as seasonality, geographical dispersion, demand uncertainties, temperature sensitivity and cold chain requirements, the perishable nature of products, short shelf life, and agricultural policies make supply chain network design particularly challenging. By its very nature, supply chain network design ensures that products reach consumers at the right place, at the right time, and at the minimum possible cost. Given these circumstances, the agri-food sector necessitates the development of integrated decision support systems that aim to optimise production and sales while improving resource utilisation efficiency in an increasingly competitive environment. In this study, a multi-stage, multi-node supply chain network model that maximises sales profit was developed using real data from a food company producing garlic in the Taşköprü district of Kastamonu, taking into account all the challenges and complexities of the sector. In this study, the first step was to optimise the Vehicle Routing Problem with Time Windows (VRPTW) in order to construct the sales component of the supply chain network (SCN), thereby minimising the costs of distribution channels. A mathematical model was developed for the VRPTW and solved using the GAMS program, through which distribution routes were generated while simultaneously minimising time, number of vehicles, distance, and costs. The routes obtained in this stage, along with their associated costs and demand quantities, were then used as input data for the main focus of the study, Supply Chain Network Design (SCND). In the second part, the focus was on SCND, the main subject of the research. In this stage, the routes and their costs obtained from VRPTW were used as data to maximise sales profit, and a mathematical model was developed. This model, solved using the GAMS program, incorporated constraints related to capital balance, periodic production, contract manufacturing, external procurement, inventory management, waste, and fixed costs. Unlike classical production planning models, this model included unique features such as dynamic decision structures based on point-of-sale benefit–cost analysis and waste recovery strategies. The study also carried out sensitivity analyses on production, contract manufacturing, inventory capacity, and capital. The scenarios revealed that sales profit is significantly affected and, in certain cases, bottlenecks occur. In conclusion, the study highlights the need for integrated decision-making in production, procurement, storage, sales, and capital planning within supply chain network design for the agri-food sector. It demonstrates that integrated and optimised decisions can enhance long-term sustainability and profitability. The proposed supply chain network model offers a strategic framework to guide decision-making in the agricultural food sector.