Optimization Under Uncertainty in Sustainable Agriculture and Agrifood Industry

Chile is currently the fourth-largest wine-exporting country in the world. Most of the wine is transported in dry containers, exposing it to the prevailing temperature conditions during its maritime transport, which can affect its quality. Transport decisions are mostly based on costs, with the least cost route being preferred usually, without considering the potential temperature risks. In this study, we develop a decision support model for the shipping route selection problem, considering the temperature risk during maritime transport. To achieve this, we construct a model that considers the internal container temperature information obtained from 167 shipments of wine and determines the correlation with the external temperature. Because the external temperature is available through the global NCEP-NCAR database, we can determine the internal container temperature of any shipping route. We also present a set of temperature risk indices, which allows us to assess the risk to the wine shipment for a specific route. The results indicate a good forecasting performance for our model, with low mean accumulated deviation and root mean squared error values. We validate this model by applying it to a group of routes and show that the lowest-cost route can have the highest risk for wine quality. Hence, a more expensive and less risky alternative route should be considered.

The state of Mato Grosso, Brazil’s largest grain producer, presents a logistical deficit that reduces its competitivity for having a unique rail solution that is located in its state, “Rumo Malha Norte” (RMN) transhipment terminal in Rondonópolis (MT). In that way, the railway system expansion presents a major role for logistic cost reduction. The objective of this paper was to present a comparative analysis for the current and prospected corridor solutions for Mato Grosso’s state grain exportation market, considering the demand calculation for each railway and the impact on total cost reduction in order to define the best investment option for the public and private actors. For achieving the results, it was structured a six-step algorithm that considered roadway freight prices and had the scenarios built upon discounts for composing the rail transportation prices. In terms of cargo capturing, Ferrogrão railway has the highest potential for the state, although RMN expansion to the municipalities of Cuiabá (MT), Nova Mutum (MT), or Lucas do Rio Verde (MT) also presents good results. Furthermore, the best solutions (Ferrogrão and all three RMN expansions) were compared, and Ferrogrão project is also more sensible to reductions in rail freight prices over roadway transportation, leading to lower total costs for the actors in the market.

In our contribution, we address advanced problems of plant variety experiments, which are organised at the official level of the Czech Republic by the Central Institute for Supervising and Testing in Agriculture (CISTA), a specialised authority established by the Ministry of Agriculture of the Czech Republic. Experiments carried out by CISTA are typically conducted with a low number of replications, and it is often necessary to take into account the limited experimental area when planning an experiment. These are mostly organised according to Alpha-design which represents a specific optimisation issue. In official plant variety testing, various restrictions appear (shape of the area, height of plant varieties, appearance of the replicated neighbourhood of the plant variety) which are not included in the original optimisation, and ex post design modifications must be made. The aim of this chapter is to describe the process of the development and validation of a decision support system (DSS) that accommodates an optimal Alpha-design for these restrictions and provide detailed information on the implementation of the DSS in practice. The DSS was developed in 2012–2013 and implemented through 2014–2015, but the development of the DSS is still ongoing. Currently, it is in routine usage and works effectively.

Food products of animal or vegetal origin constitute one of the most important business sectors worldwide. Providing appropriate nutrition to a growing world population with ever-increasing dietary habits is a major challenge current practitioners face. To ensure the sector can successfully respond to present and future challenges, the appropriate management of agri-food supply chains (AFSC) is mandatory. This work explores this challenge and proposes a new modelling approach specific to support AFSC design and planning decisions. Our approach uses a mixed-integer linear programming (MILP) model to design a quantitative model adapted to the context of AFSC, exploring product perishability, technological processing alternatives, different storage capacity strategies, and reverse logistics under an environment where uncertainty in supply and demand exists. The model is applied in a case study, primarily drawn from an existing sugar beet supply chain in the Netherlands. The results show that there is space for better management of AFSC, namely, taking advantage of closed-loop SC principles. Finally, future research directions are highlighted to support further investigation in this field.

Agricultural supply chains for fresh products exhibit higher levels of variability than those exhibited by other agri-food supply chains for products with longer shelf lives such as grains, cotton, or frozen foods. This variability takes different forms; chief among them are price and yield variability. These factors complicate the task of the grower of deciding when and how much of each crop to plant to maximize expected profits. In this chapter we discuss mathematical models for making planting decisions that are robust to the price and yield variability of fresh foods so that growers can maximize their expected profit. As part of the chapter, a case study is presented. The case study allocates production among complementary regions using a stochastic programming approach.

Agriculture plays an essential role in sustaining human activities. Challenges such as the indiscriminate use of pesticides pose a threat to food security. Evolutionary computing (EC) has emerged as a robust computational methodology for the treatment of many complex agricultural problems in recent years. In addition, scientific workflows are a technology that supports the automation and reproducibility of large-scale in silico experiments. However, the design of evolutionary workflows is still an open issue for decision-makers. Therefore, to bridge this gap, we present a novel approach to help researchers model evolutionary workflows. To answer this question, in this chapter, we use VisPyGMO, which offers a set of evolutionary algorithm modules that help researchers build reusable evolutionary workflows more efficiently. Moreover, we show the feasibility of VisPyGMO in analysing a large real-world agricultural dataset used to respond to competency questions (CQ) and predict future use of pesticides.

This work tackles the management zone delineation and the crop rotation planning problems in an integrated precision agriculture framework. The zoning problem allows to define relatively homogeneous management zones regarding their soil properties and for which specific rates of agricultural inputs are necessary. From a sustainable point of view, cropping of species from different botanic families in adjacent zones at the same time can aid pest control and a reduction in the use of pesticides. More specifically, we propose a novel linear binary integer program for an integrated zoning and crop rotation planning problem with adjacency constraints. In this model, we maximize the incomes of the crop rotation plan subject to zoning constraints and adjacency constraints on crop families. The proposed model is solved using a column generation method with column-dependent rows as an algorithmic strategy. Results from a set of instances show the relevance of the proposed model as well as the need to have an algorithmic technique for solving the proposed model, being able to find acceptable feasible solutions to medium- and large-sized problems that cannot be solved directly by a commercial software.

Nowadays, many fresh fruit companies choose produce suppliers and the subsequent fresh produce purchase arbitrarily. They do not consider the interaction between fresh fruit quality and the required cold storage technology. These decisions are critical because they have a great impact on final product quality and process efficiency. Since fruit harvesting is seasonal, the selection of suppliers and required refrigeration technology choice are made before or during the harvest season. However, many uncertainties arise such as market conditions, climate, exchange rates, and raw material shortage, among others. These dilemmas make it necessary to react quickly and flexibly when facing them. Therefore, a decision support system based on a mathematical programming model was developed in this study for supporting these tactical decisions. This system was named the FruitPS-DSS and was applied in a Chilean dehydrated fruit company to validate the system functionality. In this company, an analysis of different possible scenarios concerning changes in demand and fruit price during a supply-planning season was performed. Moreover, it was deployed and validated according to the experience of three operational managers who were in charge of the fruit purchase and storage processes of their respective companies. They positively valued the obtained reports and information, mentioning that it will allow them to make timely decisions under a user-friendly interface. Additionally, they pointed out that the FruitPS-DSS could facilitate ordering and handling information. Finally, it is important to notice that the FruitPS-DSS is flexible enough to be adapted to different kinds of fruit processing plants.