Abstract
This work aims to integrate Remote Sensing (RS) and cadastral data in QGIS software to perform the spatiotemporal mapping of Wine Yield (WY) cluster zones in the Douro region. Spatiotemporal modelling approach for prediction of wine yield was based on Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST) and topographic data. The results showed that 74% (R2 = 0.744, n=128, p<0.000) WY interannual variability at administrative division could be explained by the developed model. This information allows establishing wine production region pattern which can improve the agronomic and economic efficiency of vineyard and winery operations.

Abstract

Abstract
The usage of unmanned aerial vehicles (UAVs) in civil and military applications continues to increase due to the numerous advantages that they provide over conventional approaches. Despite the abundance of such advantages, it is imperative to investigate the performance of UAV utilization while considering their design limitations. This paper investigates the deployment of UAV swarms when each UAV carries a machine learning classification task. To avoid data exchange with ground-based processing nodes, a federated learning approach is adopted between a UAV leader and the swarm members to improve the local learning model while avoiding excessive air-to-ground and ground-to-air communications. Moreover, the proposed de-ployment framework considers the stringent energy constraints of UAVs and the problem of class imbalance, where we show that considering these design parameters significantly improves the performances of the UAV swarm in terms of classification accuracy, energy consumption and availability of UAVs when compared with several baseline algorithms.

Abstract
This paper investigates the relationship between students' performance and the type of school attended during upper secondary education. The performance of three different types of schools (Liceo, Technical and Professional schools) in four Italian macroregions (North West, North East, Centre, South & Islands) is investigated. A benchmarking analysis of the variability in students' performance among regions (within macroregions) for cohorts of students attending Liceo is also conducted. The data was collected at the student level from the Italian Institute for the Evaluation of Education System (INVALSI), for the academic year 2017/18. Families with higher socio-economic status may self-select into Liceo, so a direct comparison with vocational schools could lead to biased conclusions regarding the impact of school type on student performance. To overcome this limitation, we used a Propensity Score Matching approach prior to the estimation of efficiency. A pseudo-Malmquist index, based on a metafrontier and satisfying the circular property, is developed. It enables comparing the location of the best-practice frontier for each type of school and the spread in the educational efficiency of the students attending each type of school. Thus, best performance of a given school type corresponds to the combined effect of these two aspects. This study is an interesting starting point to challenge the stereotypes that persist in Italy, especially concerning general and vocational studies and geographic differences in educational achievements.

Abstract
The storage policy has a tremendous impact on the efficiency of order-picking operations, which can account for up to 50% of operating costs. The coronavirus pandemic has reinforced the importance of managers making the right operational decisions, namely regarding the definition of the storage policy. It is therefore important to consider handling constraints. This article is inspired by a Portuguese retail company and it considers two handling constraints: weight and shape. We define the location of products by using a zero-one quadratic assignment model. In this model, in addition to the demand and similarity, we considered the weight and shape of the products. We used both weight and shape parameters to set products with similar shapes together, placing aside products with odd shapes. Our analysis shows that the inclusion of the shape and weight into the problem improved the current operations. We found that our method allowed for a reduction of up to 24% in the picking distance, a percentage higher than the one that only considers weight constraints. The inclusion of the shape parameter into the study enabled the company to increase the flow and efficiency of the order-picking operations. Thus, it can be an asset for other warehouses.

Abstract
This paper proposes a multi-objective mixed-integer linear programming to model a cold chain with complementary operations on a hierarchical hub network. Central hubs are linked to each other in the first level of the network and to the star network of the lower-level hubs. As for a case study, different hub levels provide various refreshing or freezing operations to keep the perishable goods fresh along the network. Disruption is formulated by the consideration of stochastic demand and multi-level freshness time windows. Regarding the solution, a genetic algorithm is also developed and compared for competing the large-sized networks.