https://tede.unioeste.br/handle/tede/527 2026-06-19T19:04:42Z https://tede.unioeste.br/handle/tede/8356 Title: Fontes de silício e corretivos na agricultura: efeitos sobre os atributos químicos e físicos do solo e na produtividade do trigo, milho e cana de-açúcar. Autor: OLIVEIRA, Daiane de Deus Primeiro orientador: Maggi, Marcio Furlan Abstract: The growing global concern over increasing food demand has led to the expansion of new agricultural areas. However, environmental issues—such as the reuse and recycling of waste discarded by industries—have prompted reflection on how to boost agricultural productivity while ensuring both economic gain and environmental sustainability. In this context, new technologies have been refined and implemented in the agricultural sector with the goal of reducing production costs and increasing crop yields. One notable example is precision agriculture, which uses tools such as electrical conductivity measurements combined with chemical and physical soil analyses. These techniques contribute to sustainable and balanced management practices, preventing drift and leaching of many agricultural inputs, as well as ensuring the correct application of fertilizers. For this reason, the first chapter of the thesis evaluated whether there was any interference in the chemical and physical properties of the soil in relation to electrical conductivity and the productivity of wheat and corn crops after the application of limestone and silicate. This phase of the study was carried out on a commercial agricultural property located in the municipality of Céu Azul, in the state of Paraná, Brazil. After soil collection and chemical analysis, application rates were established at 0, 300, and 800 kg·ha⁻¹ for the control, silicate, and limestone treatments, respectively. To measure soil electrical conductivity, an EM38-MK2 conductivity meter was used. Data cleaning, interpolation, and correlation analyses were conducted using the AgDataBox platform. Results showed that only the nutrient calcium presented a significant difference between the silicate and control treatments, based on the chemical sampling conducted in 2022. Regarding the correlation matrix, autocorrelation among chemical elements was observed; however, there was no correlation between electrical conductivity and magnesium in the 2024 crop season. Additionally, autocorrelation was found between soil compaction resistance and different depths, soil texture, and yield levels. A significant difference was also observed in soil compaction resistance between the control and silicate treatments. It was concluded that both positive and negative correlations, as well as autocorrelations, existed between the parameters analyzed and soil electrical conductivity. In the second chapter of the thesis, the objective was to analyze the use of different silicon sources in sugarcane production in the state of Louisiana, United States of America. The experiment was conducted at the Louisiana State University Agricultural Research Station. Fertilization treatments consisted of the application of 0 kg·ha⁻¹ for the control, 2471 kg·ha⁻¹ for the limestone treatment, 2471 kg·ha⁻¹ for the sugarcane bagasse ash treatment, and 1235.5 and 2471 kg·ha⁻¹ for the silicate slag treatments. The experimental design used was a randomized block design, with plots consisting of three rows measuring 30.48 meters in length, using the sugarcane cultivar L01-299. Soil and leaf samples were collected 30 days after fertilization and again at harvest. In the laboratory, soil electrical conductivity analyses and colorimetric determinations were performed to quantify silicon in both soil and plant tissue. It was observed that soil electrical conductivity values were higher 30 days after fertilizer application compared to those measured at harvest time. The soil silicon analyses showed similar results between the two sampling periods, with no significant differences among the fertilization treatments. The treatments evaluated 30 days after fertilization exhibited higher silicon concentrations in the plants compared to those measured at harvest. In conclusion, soil silicon levels remained relatively consistent across the different sampling times; however, the highest concentration, 182.28 mg·kg⁻¹, ix was obtained 30 days after fertilization in the treatment with 2471 kg·ha⁻¹ of silicate. Regarding silicon content in the plant, the highest level (2.50 mg·kg⁻¹) was observed in the limestone treatment 30 days after application, while a lower value (0.86 mg·kg⁻¹) was recorded for the same treatment at harvest. Overall, it can be concluded that sustainable management practices combined with the use of technological tools lead to increased productivity and profitability for producers, while also ensuring the preservation of a clean and balanced environment for future generations. Publisher: Universidade Estadual do Oeste do Paraná Tipo do documento: Tese 2025-09-04T00:00:00Z https://tede.unioeste.br/handle/tede/8337 Title: Crédito de Carbono com Prova de Reserva a Partir de Propriedades Rurais como Mecanismo de Mitigação de Emissões Globais em Conformidade com a Regulamentação de Desmatamento-EUDR e Mercados Regulados de Crédito de Carbono Autor: Ferronato, Marcos Alexandre Fernandes Primeiro orientador: Uribe Opazo, Miguel Angel Abstract: The growing urgency of the climate crisis poses significant challenges to the agricultural sector, demanding innovative solutions that reconcile productivity and sustainability. In this context, this thesis aimed to develop and validate an integrated and technologically advanced model for generating carbon credits on Brazilian rural properties, combining cutting-edge technologies with sustainable and regenerative management practices in accordance with regulated market standards, to promote the decarbonization of the economy, the valuation of ecosystem services, and climate justice in the agricultural sector. This thesis was structured in the form of articles. The first article investigated the feasibility of using sensors, artificial intelligence, and blockchain to transform rural properties into positive climate assets, generating traceable and secure carbon credits. The methodology involved analyzing data from four rural properties, focusing on quantifying carbon retention in the soil and biomass. The results demonstrated the accuracy of the technologies employed and the potential of forest and agricultural areas as legitimate sources of carbon credit. The second article aimed to demonstrate the technical and legal feasibility of generating carbon credits on Brazilian rural properties, emphasizing traceability, security, and tokenization of credits. The methodology involved analyzing geospatial data, remote sensing, and applying convolutional neural networks. The results indicated that, with the proper use of emerging technologies and in compliance with legal frameworks, rural properties of different sizes can consolidate themselves as safe and economically viable environmental assets. The third article proposed an integrated model to quantify net negative emissions on rural properties with native forests, aiming to generate carbon credits and promote climate justice. The methodology combined artificial intelligence, georeferenced data, and environmental auditing principles. The results demonstrated that the proposed model can generate carbon credits and promote climate justice, adequately remunerating countries with positive climate performance. In conclusion, this thesis demonstrates the feasibility of an innovative and integrated model for generating carbon credits on Brazilian rural properties, which aligns with regulated market standards. The combination of advanced technologies, sustainable management practices, and the valuation of ecosystem services presents a promising path to promote the decarbonization of the economy, the valorization of the agricultural sector, and climate justice, contributing to a more sustainable and resilient future. Publisher: Universidade Estadual do Oeste do Paraná Tipo do documento: Tese 2025-02-28T00:00:00Z https://tede.unioeste.br/handle/tede/8291 Title: AgDataBox-Map 5.0: modernização do front-end – migração de Angular para React com foco em desempenho, manutenibilidade e paridade funcional Autor: Bellincanta Filho, Nelson Primeiro orientador: Souza, Eduardo Godoy de Abstract: The modernization of AgDataBox-Map (AgDataBox-Map) was carried out to enhance maintainability, performance, and user experience quality in a web-based geospatial application focused on Precision Agriculture. The work involved migrating the front end from Angular to React (with Vite and TypeScript), preserving stable contracts with the ADB-API and ensuring functional parity with the legacy version through an incremental, route/module-based strategy, supported by objective acceptance criteria and immediate rollback capability. The evaluation followed a controlled before-and-after experimental design, combining verification of equivalence across critical workflows (authentication, projects, data import/cleaning, spatial operations, and the management zone pipeline), build and startup metrics, bundle size variations, as well as exploratory inspections of usability and essential accessibility. The results indicate the maintenance of functional parity in the evaluated scenarios and perceived gains in responsiveness and navigation fluidity, accompanied by a more modular and testable architectural foundation (React-based componentization, state management with Redux Toolkit, and an updated OpenLayers version). As an additional contribution, the study incorporates into AgDataBox-Map 5.0 the implementation of the Genetic Algorithm for Zone Delineation (GAZD) for the generation of Rectangular Management Zones (RMZs), integrating it into the existing pipeline through preprocessing by Area Units (AUs), exclusion of no-data values, and the use of variance and relative variance metrics to control homogeneity.Finally, the thesis is structured as follows: (i) Section 2 presents the general and specific objectives; (ii) Section 3 is dedicated to the literature review on front-end development and architecture, migration strategies, and UX foundations applied to geospatial systems; (iii) Section 4 details the materials and methods (scenarios, metrics, testing instruments, and the equivalence protocol); (iv) Section 5 presents the results and discussion; and (v) Section 7 consolidates the conclusions and outlines directions for future work. Publisher: Universidade Estadual do Oeste do Paraná Tipo do documento: Tese 2025-11-11T00:00:00Z https://tede.unioeste.br/handle/tede/8280 Title: Transformação digital de processos, produtos e serviços da empresa sob a ótica da Indústria 4.0. Autor: Freire Filho, Fernando Celio Monte Primeiro orientador: Johann, Jerry Adriani Abstract: Digital transformation has established itself as a primary driver of organizational competitiveness, redefining production processes, management models, and business strategies. This doctoral thesis investigated how process standardization, managerial excellence, and the systematic use of data accelerate organizational maturity toward Industry 4.0 in a Brazilian agribusiness company undergoing modernization. The study was conducted within the scope of the Academic Doctorate for Innovation (DAI/CNPq), in partnership with the analyzed organization, and included international cooperation through the PDSE/CAPES program. The study adopted an applied, mixed-methods approach, combining: (i) documentary analysis; (ii) structured managerial diagnosis; and (iii) analytical modeling of corporate data. The research design integrated two complementary dimensions: management consolidation and process standardization (supported by ISO 9001:2015 and the Management Excellence Model – MEG); and the incorporation of analytical methods and intelligent decision-support systems (CRISP-DM process integrated with Business Intelligence). In the managerial dimension, ISO 9001 certification provided the governance foundation and evidence for maturity assessment via MEG, demonstrating statistically significant improvements in key criteria. Notable advances were observed in Leadership (mean 37.5 – 74.5, p = 0.001), Strategies and Plans (31.3 – 58.3, p = 0.003), and Customers (45.0 – 82.5, p = 0.008), evidencing a strengthened strategic direction, process discipline, and focus on customer value. These results confirm that quality management acts as the organizational infrastructure for digital transformation by reducing variability, increasing traceability, and institutionalizing continuous improvement. In the technological-analytical dimension, a complete six-phase CRISP-DM pipeline was implemented using the 2018–2023 historical series, resulting in three BI management dashboards (sales, customers, and sales force) and a framework of prioritized insights for commercial action. The integration of operational and commercial data structured Key Performance Indicators (KPIs) and enabled continuous monitoring and comparative evaluation by period, product mix, and portfolio, thereby expanding the organization’s capacity for learning and evidence-based decision-making. As a practical outcome, the company began identifying seasonality patterns, revenue concentration, and regional/customer opportunities in actionable time, shortening diagnostic cycles and directing efforts toward high-contribution segments and items. In summary, findings demonstrate that the journey toward Industry 4.0 is evolutionary and cumulative: it begins with the consolidation of quality and process governance (ISO 9001 + MEG, showing quantitative gains in means and p-values) and matures through business analytics (CRISP-DM + BI, with defined temporal scope, delivered artifacts, and monitoring routines). The combination of standardization, management, and technology resulted in a data-driven organization where processes, products, and services are redesigned based on efficiency and connectivity logic. The study concludes that the synergy between management maturity and analytical competence consolidates a digital transformation model that is both applicable and sustainable within the Brazilian agribusiness context. Publisher: Universidade Estadual do Oeste do Paraná Tipo do documento: Tese 2025-11-07T00:00:00Z