Israel David Herrera Granda

Facility layout planning (FLP) in the agri-food sector faces unique complexities due to product perishability, biosafety regulations, and Industry 5.0 (I5.0) requirements. The reviewed literature is skewed toward static formulations that overlook detailed layout, demand uncertainty, and social sustainability. In response, a targeted review, reported following PRISMA 2020, and expert judgement yielded an ontological classification of ten characteristics influencing detailed agri-food layouts, whose relevance and relationships were supported by five external experts (S-CVI/Ave = 0.96); the system roles proposed by the authors were not reproduced in a blind assignment and remain modelling hypotheses. We then develop an eight-pillar conceptual framework addressing dynamic multi-period planning, stochastic uncertainty, multi-objective optimisation, comprehensive sustainability, human-centric ergonomics, operation-level routing, a bottom-up approach, and perishability as an endogenous variable. A diagnostic case study of an agri-food small and medium-sized enterprise (SME) illustrates how the classification structures the diagnosis of layout inefficiencies, such as products reaching customers with reduced shelf life. Aligned with I5.0 principles, this research provides a conceptual basis for models that treat perishability and human well-being as core design variables. The classification applies mainly to agri-food SMEs operating under make-to-order and just-in-time strategies; large make-to-stock manufacturers should also consider product homogeneity and storage-based product types.

Ecuadorian agri-food supply chains are characterized by strong territorial heterogeneity, crop perishability, and unequal production concentration, creating significant socio-economic and logistical challenges. To address these territorial disparities, this study employs an AI-driven clustering approach to classify Ecuador’s 23 continental provinces by their agri-food production and supply chain integration profiles. Using official 2024 provincial tabulations from the National Institute of Statistics and Censuses of Ecuador, seven variables were analyzed: harvested banana area, banana production, cocoa production, rice production, corn production, potato production, and banana yield. The methodology combined descriptive statistics and Pearson correlations with unsupervised machine learning techniques, specifically hierarchical agglomerative clustering and k-means algorithms after z-score standardization. The descriptive results showed extreme dispersion across provinces, with coefficients of variation exceeding 100% for all variables and marked right skew. The final four-cluster solution (k = 4, BSS/TSS = 78.65%, average silhouette = 0.43) differentiated: (cluster 1) a potato-specialized Andean profile represented by Carchi; (cluster 2) a broad group of low-intensity provinces requiring inclusive logistical integration; (cluster 3) medium-scale banana provinces with high yields; and (cluster 4) the dominant tropical production core formed by Guayas and Los Rios. The typology explains 78.65% of the standardized variability and provides the territorial evidence base required for designing differentiated supply-chain planning, resilience strategies, and public policies tailored to each provincial typology.

This work consists of a systematised review examining recent models for agri-food supply chain management (AFSCM), focusing on artificial intelligence (AI) applications and sustainability integration. Through PRISMA methodology, we conducted a descriptive statistical analysis of 183 articles published between 2019 and 2023. The dimensional analysis revealed that strategic modelling represented 55% of approaches, while perishable products were considered in 58% of studies. Sustainability principles were integrated into 74% of the models, while AI applications showed a limited 14% adoption across AFSCM. The analysis of AI implementations showed distribution in agricultural production management at 27%, transportation and logistics at 8%, and sustainability planning at 19%. A novel AFSCM taxonomy was developed through systematic classification, combining quantitative bibliometric analysis, qualitative thematic analysis and expert interpretation. The results identified gaps in mixed methods approaches and tactical or operational modelling, providing a structured framework for future research directions in AI integration and sustainable practices across agri-food supply chains.

In recent years, SLAM, visual odometry, and structure-from-motion approaches have widely addressed the problems of 3D reconstruction and ego-motion estimation. Of the many input modalities that can be used to solve these ill-posed problems, the pure visual alternative using a single monocular RGB camera has attracted the attention of multiple researchers due to its low cost and widespread availability in handheld devices. One of the best proposals currently available is the Direct Sparse Odometry (DSO) system, which has demonstrated the ability to accurately recover trajectories and depth maps using monocular sequences as the only source of information. Given the impressive advances in single-image depth estimation using neural networks, this work proposes an extension of the DSO system, named DeepDSO. DeepDSO effectively integrates the state-of-the-art NeW CRF neural network as a depth estimation module, providing depth prior information for each candidate point. This reduces the point search interval over the epipolar line. This integration improves the DSO algorithm’s depth point initialization and allows each proposed point to converge faster to its true depth. Experimentation carried out in the TUM-Mono dataset demonstrated that adding the neural network depth estimation module to the DSO pipeline significantly reduced rotation, translation, scale, start-segment alignment, end-segment alignment, and RMSE errors.

In recent years, SLAM, visual odometry, and structure from motion approaches have widely addressed the problems of 3D reconstruction and ego-motion estimation. Of the many input modalities that can be used to solve these ill-posed problems, the pure visual alternative using a single monocular RGB camera has attracted the attention of multiple researchers due to its low cost and widespread availability in handheld devices. One of the best proposals currently available is the Direct Sparse Odometry (DSO) system, which has demonstrated its ability to accurately recover trajectories and depth maps using monocular sequences as the only source of information. Given the impressive advances in single-image depth estimation using neural networks, this work proposes an extension of the DSO system, named DeepDSO. DeepDSO effectively integrates the state-of-the-art NeW CRFs CNN as a depth estimation module, providing depth prior information for each candidate point. This reduces the point search interval over the epipolar line. This integration improves the DSO algorithm's depth point initialization and allows each proposed point to converge faster to its true depth. Experimentation carried out in the TUM-Mono dataset demonstrated that adding the CNN depth estimation module to the DSO pipeline significantly reduced rotation, translation, scale, start-segment alignment, end-segment alignment, and RMSE errors. The DeepDSO code is publicly available at https://github.com/erickherreraresearch/Deep-DSO.

Bactericera cockerelli (Sulc) is the main vector of plant phloem-associated pathogens responsible for the potato purple top complex, a disease that can cause losses of up to 100% in potato crops. Chemical control methods, mainly based on insecticides, have proven to be inefficient due to the development of insect resistance. This problem highlights the need to implement sustainable strategies to mitigate its impact. One promising strategy is the genetic resistance of plants, specifically antibiosis, this type of resistance implies that plants negatively affect the insect's biology. In this context, the present study aimed to identify potato genotypes conserved by the National Roots and Tubers Program of the National Institute of Agricultural Research with antibiosis resistance characteristics against B. cockerelli, by means of a statistical analysis based on a complete factorial design. The evaluation considered the number of eggs oviposited on different days of evaluation in 9 wild species, 18 improved varieties, 15 native varieties and 18 promising clones. Genotypes with a marked resistance due to antibiosis were identified. These materials can contribute valuable genetic resources for breeding programs and the implementation of integrated crop management strategies, thus reducing dependence on pesticides that endanger human health and biodiversity.

The difficulty in learning of the Automotive Electrical and Electronic Systems training module by the students from the Third Technical Baccalaureate in Automotive Electromechanics, leads to the present research has as the aim to implement the use of Augmented Reality (AR) in the process of teaching. The applied research was documental, field, and quasi-experimental type, with a mixed approach. A pre-test was engaged to 38 students from two parallels of the General Julio Andrade Educational Unit, located in the province of Carchi (Ecuador), to identify the prior knowledge, and a post-test after of using AR to know the level of acquired learning. With the development of the established proposal, an improvement in academic performance was evident, and it was reflected on the obtained result in the post-test in the students of parallel B with an average of 8,50/10 points, unlike the score from the parallel A, in which the traditional methodology was applied, and 7,78/10 points were gained. As a conclusion, the application of the technological tool of AR is feasible and generates in the students a high interest of learning in the aforementioned module.

The assessment of learning achievements is a topic of great interest and has been widely addressed in educational institutions worldwide. Especially in the university context, it is a topic that has been widely addressed from different perspectives using quantitative and qualitative tools. This study presents the evaluation model for the achievement of learning outcomes in the Food Engineering career of the State Polytechnic University of Carchi, which is carried out at the end of each academic period by applying a structured test in each subject at each level. The data addressed in this study correspond to the academic period 2022A, which was collected in a database and processed using a proposed protocol of statistical tests composed of Mahalanobis distances as data processing technique and Kruskal-Wallis and Dunn- Šidák tests as tests for differences. The results show that the data collected through structured tests for the evaluation of learning achievements are susceptible to the presence of atypical observations and generally present a non-parametric behavior, so the application of data processing techniques before inferential analysis it's of vital importance. Furthermore, it has been found that not all subjects or levels tested show significant differences so subjects showing lower performance should only be classified as having learning problems if the differences observed by a statistical test are significant. In addition, the use of mean, median, mode, and trimmed mean as metrics for comparing attainment was compared and it was found that the most appropriate metrics for this task were the median and the trimmed mean.

Reducing the carbon footprint of higher education institutions has become a global trend in recent years, as they are the axis of society's scientific and technological transformation that must be carried out with a sustainable approach. Among the many approaches that can contribute to this problem are the digital transformation of academic and administrative processes and its impact on reducing paper use. This study details the statistical analysis of the digital transformation process based on the modules and computer services implemented by the ICT department in a higher education institution and its incidence in reducing the use of paper by its teaching and administrative staff. For this purpose, an initial instrument was proposed, consisting of 18 ordinal, three categorical and two numerical variables. After applying statistical data processing and validation techniques, it was reduced to 13 ordinal variables, three categorical and two influential numerical variables. In addition, this study has proposed an appropriate statistical protocol for the design and extraction of information applicable in the context of higher education institutions, consisting of the techniques: Mahalanobis distances, confirmatory factor analysis, Kruskal-Wallis, Mann-Whitney's U and Wilcoxon tests. The results showed the power of confirmatory factor analysis as a technique for obtaining models to measure indirectly observed latent variables, such as perceptions of digital transformation and paper reduction. In addition, it was found that the modules, systems, and computer tools implemented in the institution significantly reduced the execution times for performing administrative and teaching tasks in the institution, and significant differences were identified in the perception of digital transformation and reduction of paper use among staff.