The goal of the awards, as stated in the call conditions, is “to advance the efforts of researchers of Spanish nationality working in Statistics and Operations Research,” and, by recognizing excellence in these two disciplines, “to support their transmission to society at large.” The accolades in this edition have gone to five papers produced by researchers at universities and research centers in Madrid, Catalonia, Valencia, Andalusia, Galicia and Navarra, with the collaboration of scientists from other European countries including Italy, Portugal, Belgium, Denmark and Hungary. Their award-winning contributions have been published in high-impact international journals, driving new theoretical and methodological advances of widespread significance.

Statistics deals with data analysis, the relating of mathematical models to reality, while the aim of Operations Research is to optimize decision-making. Nowadays vast amounts of data are generated in every area of research, but in order for these data to be processed into useful information our increasingly powerful computers must be equipped with the right mathematical tools. Hence the growing centrality of Statistics and Operations Research, the disciplines that provide the tools needed for data analysis across all branches of science, and which underlie many of the advances we now take for granted, from Internet search engines to our smartphone apps.

The awards, funded with 6,000 euros in each of their five categories, are bestowed for excellence in scientific contributions published in the last five years. Their authors must be researchers of Spanish nationality, or those of other nationalities that have conducted their investigation in a university or scientific center in Spain. Awards may also go to persons of any nationality for contributions made in collaboration with one or more Spanish nationals.

This partnership between the BBVA Foundation and SEIO to showcase the talent of Spanish researchers in Statistics and Operations Research joins three more annual award programs that the Foundation co-organizes with Spanish scientific societies; namely the Physics Awards with the Spanish Royal Society of Physics (RSEF), the Vicent Caselles Mathematical Research Awards with the Spanish Royal Society of Mathematics (RSME), and the Computer Science Research Awards with the Spanish Society of Computer Science (SCIE).

AWARDEES

Best methodological contribution in Statistics

María Alonso Pena, Assistant Professor of Statistics and Operations Research at the University of Santiago de Compostela; Irène Gijbels, Professor of Statistics at KU Leuven (Belgium); and Rosa Crujeiras, Professor of Statistics and Operations Research at the University of Santiago de Compostela, receive the award for the best methodological contribution in Statistics for their paper “A General Framework for Circular Local Likelihood Regression,” published in the Journal of the American Statistical Association (JASA).

This article proposes a general framework to explain the effects of certain directional statistical variables on a variable of interest. For instance, to study how pollutants will behave as a function of wind direction, it is important to ascertain the angle the wind is blowing from. Likewise, when analyzing neural response to visual stimuli, the direction they come from is vital data that may affect how the neurons behave. And, by the same token, the flight strategy of prey will vary according to the angle from which a predator is approaching.

In all these contexts, directionality is a key input to statistical analysis, but existing proposals for processing this kind of data exhibit a series of shortcomings from both a theoretical and a practical standpoint. The winning team have put forward a more general framework, one that is “theoretically grounded,” as Crujeiras puts it, and whose performance “has been assessed in diverse simulation scenarios to validate its use.”

Best methodological contribution in Operations Research

The award for the best methodological contribution in Operations Research is shared by Jordi Castro, Professor of Statistics and Operations Research at BarcelonaTech (UPC); Laureano Escudero, retired Professor of Statistics and Operations Research and ad honorem researcher at Rey Juan Carlos University (URJC); and Juan Monge, Professor of Statistics and Operations Research at Miguel Hernández University of Elche, for the paper “On Solving Large-scale Multistage Stochastic Optimization Problems with a New Specialized Interior-point Approach,” published in the European Journal of Operational Research.

The award-winning article develops a new mathematical optimization method for decision-making under uncertainty. When it is not possible to know with precision all the data affecting an imminent decision, stochastic optimization algorithms can be a great help. However they tend to involve multiple variables and equations that slow down the calculation process. The new method manages to run these calculations far faster than before, as Laureano Escudero explains: “We have tackled cases of up to 800 million variables and 100 million equations, and with the new method were able to solve them with one day’s calculations, while even the best of existing techniques could require weeks or even months.”

Data uncertainty arises in such varied contexts as public health, logistics or renewable energy generation, and even extends to emerging sectors like drone traffic, the compatibility of organ transplants or natural language processing. “The tool we developed allows a more efficient solution of large-scale problems (i.e., which model reality better), with the resulting savings in time, as well as in energy and CO₂ emissions,” Escudero points out.

Best applied contribution in Statistics

Jesús López Fidalgo, Professor of Statistics at the University of Navarre; Caterina May, Professor of Statistics at the University of Eastern Piedmont (Italy); and José Antonio Moler, Professor of Statistics and Operations Research at the Public University of Navarre, receive the award for the best applied contribution in Statistics for their paper “Designing Experiments for Estimating an Appropriate Outlet Size for a Silo Type Problem,” published in Annals of Applied Statistics.

Large storage facilities such as grain silos are regularly subject to blockages that are not only costly in economic terms but may also have a serious environmental impact. “Take the real case of a mine,” says López Fidalgo, “where the mineral is tipped down a vertical tunnel, and any jams that form have to be cleared with explosives.” The winning paper provides statistical tools to model jams in silos and similar storage facilities, the aim being to minimize their impact. “By accurately modeling the process,” the researcher continues, “we can work out the right diameter of tunnel to avoid jams for as many years as are needed to amortize the mine.”

In effect, the study has optimized the design of small-scale experiments to simulate jams, determining the outlet sizes to be used in experimental simulations so the model is as close a fit as possible. Hence the committee’s conclusion that the team’s contribution “has a clear social and economic impact in agriculture and industry, with important implications in risk assessment.”

Best applied contribution in Operations Research

The awardees in the best applied contribution in Operations Research category are Péter Biró, Senior Researcher at the Centre for Economic and Regional Studies HUN-REN and Associate Professor at the Corvinus University of Budapest (Hungary); Flip Klijn, Scientific Director of the Institute for Economic Analysis, CSIC, and Research Professor at the Barcelona School of Economics; Xenia Klimentova, Senior Researcher at the Centre for Enterprise Systems Engineering of the Institute of Engineering, Technology and Systems and Computer Science, INESC TEC (Portugal); and Ana Viana, Coordinating Professor in the School of Engineering of the Polytechnic of Porto and Senior Researcher in the Centre for Industrial Engineering and Management of the Institute of Engineering, Technology and Systems and Computer Science, INESC TEC (Portugal), for their paper “Shapley-Scarf Housing markets: Respecting Improvement, Integer Programming, and Kidney Exchange,” published in Mathematics of Operations Research.

The authors have come up with new mathematical algorithms to optimize paired kidney exchange programs. This strategy is used when a patient needs a kidney transplant but can find no compatible donors among their family and friends. In such cases, the patient and incompatible donor can enroll in a registry of incompatible pairs, such that matches can be found and swaps arranged with other pairs from the donor/recipient pool. “These programs,” explains researcher Flip Klijn, “have methods in place to allocate compatible kidneys,” from donors that are not known to the recipient.

In their award-winning project they tested the effectiveness of a novel game-theoretical strategy that incentivizes each patient to bring more than one donor to the paired kidney exchange program. The algorithm they designed prioritizes recipients who have recruited more potential donors to the program. “The greater the number and quality of the donors you sign up, the better your chance of getting the kidney you need,” explains Klijn. “In fact, our incentives to create a bigger and better donor pool are also good for other patients, ensuring the program works better for everyone.” The team believe their technique may serve to expand the scope and effectiveness of international kidney exchange programs.

Best contribution in Statistics and Operations Research applied to Data Science and Big Data

Emilio Carrizosa, Professor of Statistics and Operations Research at the University of Seville; Jasone Ramírez-Ayerbe, pre-doctoral researcher in Statistics and Operations Research at the University of Seville; and Dolores Romero, Professor of Operations Research at Copenhagen Business School, are recognized in the category of best contribution in Statistics and Operations Research applied to Data Science and Big Data for their paper “Mathematical Optimization Modelling for Group Counterfactual Explanations,” published in the European Journal of Operational Research.

The winning contribution provides new mathematical models for building explainable solutions into artificial intelligence algorithms. When an artificial intelligence algorithm classifies a person negatively, let’s say considering them at risk of a disease, it is often impossible to know the reasons that have led it to this conclusion. In such cases, says Carrizosa, “it would be good if this negative classification was accompanied by a contrafactual explanation; that is, the set of similar features on the basis of which the algorithm would have issued a positive score.”

To this end, the team has developed new mathematical models and optimization tools to identify the features (like lifestyle habits in the above example) that would produce a positive classification, by this means making artificial intelligence algorithms more explainable. In addition, the study considers two complementary perspectives: that of the individual being assessed, who is seeking ways to improve their personal situation, and that of a planner who, as the researcher describes it, “looks for counterfactuals for a whole community allowing for aspects such as transparency, equity or diversity.”

International committee

The international committee has a membership proposed by SEIO and the BBVA Foundation. Chairing the committee on this occasion was Daniel Peña, Emeritus Professor of Statistics at Carlos III University of Madrid (UC3M), with fellow members Michael Greenacre, Professor of Statistics at Pompeu Fabra University and the Barcelona School of Economics; Martine Labbé, Professor of Operations Research at the Université Libre de Bruxelles (Belgium); Alfredo Marín, Professor of Statistics and Operations Research at the University of Murcia; María Dolores Ruiz, Professor of Statistics and Operations Research at the University of Granada; and Carla Marina Vairetti, Associate Professor of Industrial Civil Engineering at the Universidad de los Andes (Chile).

About SEIO

SEIO is a non-profit organization whose purposes include the advancement of Statistics and Operations Research in Spain through the promotion of research, its dissemination to society, and the improvement of education at all levels. Its main goals are to communicate the quality and achievements of Statistics and Operations Research, to promote their teaching and learning, to apprise the public of the importance of both disciplines, and to serve as a reference point in all matters pertaining to science and technology.

About the BBVA Foundation

The BBVA Foundation is an expression of the BBVA Group’s engagement with the promotion of knowledge and innovation. Its activity centers on support for scientific research (through research projects, grants and collaboration with scientific institutions), the recognition of talent through families of awards organized alone or in conjunction with scientific societies, and the wider dissemination of knowledge and culture, in the conviction that fostering and relaying science-based knowledge is among the most effective means to expand our individual and collective choices. Its diverse programs, run directly or in partnership with leading institutions and organizations, focus on the areas of Basic Sciences, Biology and Biomedicine, Ecology and Environmental Sciences, Economics and Social Sciences, Statistics, Big Data and Artificial Intelligence, Information and Communication Technologies, the Humanities, Music, and the Arts.