OPTIMIZATION MODELS FOR FRESH FRUIT PRODUCTION AND DISTRIBUTION

Abstract

Vietnamese dragon fruit leads the international market, accounting for over fifty percent of total production. 20%–15% of the dragon fruit output is consumed on the domestic market, while 80%–85% is exported. Asia, Europe, and the Americas constitute the primary consumer markets, wherein China contributes between 80% and 90% of the overall annual export value. Nevertheless, the production and distribution of dragon fruit encounter various uncertainties and risks, including but not limited to weather conditions, pests and diseases, fluctuations in market prices, challenges related to logistics and infrastructure, competition, escalating input costs, sustainability concerns, and export regulations. This thesis presents research on mathematical models that aim to analyze and enhance quantitative modelling approaches utilized in the complex decision-making process of dragon fruit cultivation, harvesting, and distribution in Vietnam. By incorporating both deterministic and stochastic models, these models provide a methodical and data-centric strategy to tackle these obstacles. As a result, decision-makers are empowered to make well-informed decisions, which in turn enhance operational efficiency, minimize expenses, and improve the overall performance of dragon fruit supply chain management. The decision-making process is structured into two stages: pre-planting, when costs and resources are considered deterministic; and post-harvest, when stochastic parameters become apparent. The decision-making process is divided into tactical and operational phases using the hierarchical planning approach, which is advantageous for cultivators, producers, distributors, and vendors. The thesis has accomplished three of its objectives: conducting a comprehensive literature review on mathematical models that concern the production and distribution of fresh fruits, evaluating and developing a deterministic optimization model with certain data, and devising a stochastic optimization model to address the coupling and complex effects of uncertain factors.