چكيده به لاتين
In today’s world, agriculture is one of the most important and influential sectors in the economy of any country. The crops supply chain management is one of the main challenges of the agricultural sector due to the existence of different players, the perishability of products, the possibility of disruption, the need for balance between supply and demand, etc. In the present research, in order to plan the production and distribution of crops in an integrated manner, a single-level and bi-level nonlinear model is proposed for the perishable crops supply chain network. In this regard, the seasonal conditions of planting and harvesting, while considering disruption risk in the production sector and protection strategy to increase the reliability of the network have been considered. The goal of the proposed models is profit maximization, and in this regard, decisions are made regarding the period and amount of planting and harvesting, the protection or non-protection of farmers and how to respond to demand, so that costs such as providing seeds (seedlings), protection, packaging and transportation, Waste and wages of workers have been considered. In the single-level model, demand and price are parameters, but in the bi-level model (based on game theory), the actual demand of consumers depends on the selling price of products in the markets, and the purchase and sale price are determined based on the amount of supply and demand.
To solve the models, GAMS and MATLAB software were used, and the solution of the bi-level model was found with the help of the sequential nesting approach using the random search algorithm. According to the results of the case study conducted in this research, by removing the middlemen from the chain members and considering the direct sales by the farmers, a balance is created between supply and demand. Also, considering the disruption risk brings the model closer to the real-world conditions, and using the protection strategy, in addition to increase in profit, reduces the amount of production waste and its adverse effects. Finally, sensitivity analysis and validation have been done for the two proposed models based on the case study information.
