Identifying trade-offs in trans-continental citrus supply chains and the resulting conflicting stakeholder's incentives via physics-based, digital fruit twins

Context: Postharvest supply chains are often confronted with conflicting requirements between maximizing fruit quality and postharvest life, minimizing food loss, and minimizing energy consumption during refrigerated transportation. For example, a commonly encountered trade-off is maintaining sub-zero temperatures in a shipment to kill the larvae of fruit pests while avoiding inducing chilling injury in the fruit. In this context, the question of “what fruit attributes matter most, and to whom” comes in. Supply chains involve multiple stakeholders, including growers, exporters, regulators, distributors, retailers, and finally the consumers. Each stakeholder has different targets to meet, and therefore, the aforementioned trade-offs are assessed in different ways. This situation thereby often raises conflicting incentives for decision-making in cold chain transport. Objective: In this study, we identify the conflicting trade-offs between respiration-driven remaining quality, transpiration-driven moisture loss, mortality of fruit fly, incidence of chilling injury, risk of condensation, and environmental impact due to the energy consumed during shipping. Methods: To this end, we developed physics-based digital twins of citrus fruit in a refrigerated container. The digital twin utilizes measured delivery air temperature in commercial shipments as an input. In addition, it employs mechanistic simulations to mimic the hygrothermal and physiological fruit behavior in a shipment in-silico. We used the actionable metrics from this digital twin and translated these into desirability functions, which assess how well a combination of metrics satisfies the goals defined by the respective stakeholder. With this approach, we mapped to what extent different shipping scenarios meet the targets of the key stakeholders in the citrus export supply chain, namely exporters, regulators, distributors, retailers, and also consumers. Results and conclusions: Our findings show clear differences in how the desirability curves evolve over time for different stakeholders. We found that amongst all stakeholders, only the desirability curve of the regulator remains at a satisfactory level at the end of the supply chain. We also evaluated different shipping temperature scenarios and highlighted how the temperature trade-off manifests in different metrics. Here, we highlighted that any metric cannot be optimized in isolation without adversely influencing other metrics. Significance: Our study bridges a large gap in the quantitative estimation of stakeholder perspectives by leveraging the complementary insights provided by digital twins. This is a key step toward involving all relevant stakeholders to design the best practices and policies influencing the citrus supply chain.