Development of an intelligent platform for analysis of operational and environmental preponderant factors to obtain forest productivity (K-IA Platform)

• Grant number: 20/13772-5
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: March 01, 2022
• End date: November 30, 2022
• Field of knowledge: Agronomical Sciences - Forestry Resources and Forestry Engineering
• Principal Investigator: Ana Carolina Melo da Silva
• Grantee: Ana Carolina Melo da Silva
• Company: Kersys Desenvolvimento de Sistemas Ltda. - ME
• CNAE: Desenvolvimento e licenciamento de programas de computador não-customizáveis ; Consultoria em tecnologia da informação ; Suporte técnico, manutenção e outros serviços em tecnologia da informação
• City: São José dos Campos
• Pesquisadores principais:( Últimos ): Emanoéli de Oliveira ; Felipe Torres Sampaio ; José Roberto Andrade Pereira Junior
• Pesquisadores principais:( Antigos ): Juliana Carvalho Barbosa Ramos
• Associated researchers: José Walmir Gonçalves Duque
• Associated research grant(s): 23/06222-7 - Development of an intelligent platform for analysis of operational and environmental preponderant factors to obtain forest productivity (K-IA Platform), AP.PIPE
• Associated scholarship(s): 22/06097-5 - Development of artificial intelligence (AI) algorithms with artificial neural networks (ANN) in a platform for Big Data analysis of preponderant factors in obtaining forest productivity., BP.TT; 22/03336-9 - Development of artificial intelligence (AI) algorithms with artificial neural networks (ANN) in a platform for Big Data analysis of preponderant factors in obtaining forest productivity., BP.TT; 22/02069-7 - Development of an intelligent platform for analysis of operational and environmental prepronderant factores to obtain forest productivity (K-IA platform), BP.PIPE

Abstract

With the increase in the data volume produced by the Forestry Sector in the field and in industry, there is, in the current scenario, the need to develop intelligent tools that are capable to explore all this data properly, in the correct time and entirety, considering environmental, operational, and financial aspects simultaneously (Big Data). Tools like these, which add strategic value to existing databases in companies and optimize the decision-making process by directing the manager's performance by scientific criteria, are currently accessible only to large forest producers, who invest in the customization of exclusive tools for units, or for agricultural production. In this context, and motivated by the demand of its customers, the company Kersys Desenvolvimento de Sistemas Ltda saw the possibility of developing a new product for its portfolio: the K-IA platform, which will offer the forest manager intelligent tools to identify production bottlenecks, preponderant factors in obtaining productivity in the field and for the creation of forest planning scenarios. With this purpose, the research foreseen in this project aims to evaluate the technical and commercial feasibility of developing the K-IA platform and developing its prototype. The proposed methodology aims to achieve four main results: (1) Classification of the results of the evaluated productive areas; (2) Presentation of technical justifications of the results obtained; (3) Projection of productivity based on local operations and production conditions; (4) Evaluation and projection of financial results. These results will be achieved with: (1) the development of Machine Learning algorithms, supervised and unsupervised, which will allow the classification of results and prediction of forest productivity based on operational management and informed environmental conditions; (2) the optimization of algorithms whose development has already been started by Kersys (Decision Tree - modified ID3; Probability - Naive Bayes and Pearson's Correlation) and (3) development of the platform prototype with simplified reporting of the results of the constructed algorithms. (AU)