Areas of Concentration and Research Lines – 2025 onwards
The Graduate Program in Modeling and Quantitative Methods is part of CAPES’s Interdisciplinary Major Area and is structured around two interrelated areas of concentration:
(1) Data Science and (2) Operational Research.
Each area encompasses two lines of research, namely:
(I.a) Statistical Methods and (I.b) Computational Methods, in Data Science;
(II.a) Optimization and (II.b) Decision-Making Support Models, in Operational Research.
Although the areas of concentration and research lines of the program present differences in terms of specific purposes and techniques or even applications, there are many similarities and intersections between them, which favors complementary approaches, enhances the treatment of complex problems and provides opportunities for obtaining appropriate and effective solutions. In general, researchers in both areas rely on the same foundations and, to a certain extent, methodologies — modeling, simulation, algorithms — to generate solutions. Together, they can contribute to the advancement of scientific knowledge in these areas and produce means for companies and organizations to transform data into information, information into insights, and insights into better decisions in today’s competitive environments. By taking advantage of increasingly powerful technologies, these professionals can delve into real-world scenarios and develop creative solutions to improve everyday life.
Data Science:
Over the past two decades, we have seen an explosion in the volume of data available on the most diverse aspects of human life. The collection, storage, and processing of this data has become an extremely relevant issue in terms of providing valuable information for the treatment of modern problems. The area of data science presents itself precisely in this current and complex scenario, focusing on the discovery of structured information from large databases, usually disorganized and also populated by data with little significance, so that they can be useful for making better decisions. The methods usually employed combine statistics and computing to develop algorithms and mathematical models. The elements of this combination give rise to the two lines of research included in this area of concentration.
Statistical Methods:
One of the fundamental bases of Data Science is Statistics. The Statistical Methods research line covers this aspect of the area of concentration, working from the development of statistical models appropriate for the system or application under study to the obtaining and analysis of the solution(s) of these models, in order to assist in solving complex problems, usually affected by numerous variables. Among the most common models are those that provide predictions based on historical series, indicate trends based on scenarios, discriminate/classify new elements, situations or events based on prior knowledge.
Computational Methods: It is impossible to process a gigantic volume of data without the use of computing tools. The Computational Methods research line focuses on this aspect of Data Science. This includes several areas of activity, such as: studying efficient ways to store, organize, access and visualize large volumes of data; developing algorithms to manipulate this data and extract relevant information; processing incorrect or imprecise data, etc. Usually, such data comes from real applications in a variety of areas, such as physics, chemistry, biology, engineering, sociology, administration, and the financial market.
Operational Research:
The guiding principle of this area of concentration is the development of mathematical models and analytical methods to solve complex problems that arise in decision-making processes. The primary focus is the problem itself, its structure, characteristics and properties, and not necessarily the data associated with it. The approaches used in this area can be configured in a more theoretical bias, when the main interest lies in describing and studying the problem of interest theoretically and, from there, proposing solutions that are suited to its characteristics, or a more applied bias, when the focus is primarily on obtaining a practical solution to the target problem. In any case, they can involve several disciplines, such as mathematical optimization, statistical analysis, queueing theory, stochastic processes, analysis of complex networks, and are applied in industry, commerce, finance, business, among other sectors. The type of approach and primary focus used define the two lines of research in this area of concentration.
Optimization:
Generally speaking, this line of research is devoted to the study and mathematical resolution of problems that consist of choosing, from among a set of viable solutions, the best one, according to a comparative criterion between them. The process usually begins with the understanding of a real problem, which is then translated into one or more generic optimization problems. For the latter, mathematical models are proposed, consisting of the maximization or minimization of a function, which ranks the viable solutions, defined by a set of constraints expressed mathematically. The study of the properties of these models, the development and implementation of algorithms to solve them exactly or approximately, the evaluation of the computational efficiency of these algorithms and their application to obtain solutions to the real problem make up some of the possibilities of work in this line of research. Decision-Making Support Models: As problems and decision environments become increasingly complex, the decision-making process becomes more dependent on automated support, built using methodologies based on science and technology. This support involves the construction of models and algorithms that can handle large amounts of data and complex scenarios. The proposal, analysis, implementation and use of these tools are among the general topics of this line of research. Many problems addressed in this line of research may be optimization problems and, in this case, there is a strong intersection with approaches common to other lines of research in this same area of concentration. In other cases, the treatment of problems uses approaches from other domains, such as decision theory, game theory, probability, conflict analysis, etc.
