The mathematics are the highways of the future industrial revolution, the so-called Industry 4.0. The Barcelona Graduate School of Mathematics (BGSMath), in collaboration with Pompeu Fabra University, one of the five centres that belong to the organization, and in the framework of the activities of Mobile Week, has organized for Monday 19 February a day dedicated to the theme: “Maths for Industry 4.0”. It will be held in the Albert Calsamiglia I Blancafort hall of the Roger de Llúria building on UPF’s Ciutadella campus between 9.30am and 1.15pm.

The event will involve with the participation of mathematical research groups including the Business Analytics Research Group of the Department of Economics and Business at UPF, who are working in the areas of statistics, operational research, data science and modelling, and companies and organizations that require the digital transformation.

Extracting knowledge from information

”The increasing quantification of the industries has led to a large amount of data that may hinder the transformation of industry”, explains Arantxa Sanz, research programme manager of the BGSMath. “For many companies and industries, this is a new challenge. And it’s not a problem of computing capacity. We mathematicians offer these companies a competitive advantage because we help them to extract knowledge from information”.

Among the participants in the round table dedicated to the optimization of data analysis during “Maths for Industry 4.0”, will be Helena Ramalhinho, director of the Business Analytics Research Group at UPF and a member of the BGSMath. She is responsible for numerous projects that focus on the improvement of industrial processes and of management thanks to the development of models, algorithms and innovative methods.

Examples of optimization problems in which the group has worked are warehouse picking; distribution routes, for example for the delivery of goods by companies selling online; the location of facilities such as medical centres, or the improvement in the planning of human resources, such as the drivers of a public transport company.

“The competitive advantage that a good team of mathematicians can give a digital company might reach savings of 20 to 30% and a significant improvement in customer service”, says Ramalhinho. “We universities have the ability to innovate and invent, and the companies provide us with real problems and data on which to work”.

One of the best ways to promote cooperation between universities and industry is the creation of industrial doctorates by the Government of Catalonia, managed by the AGAUR. During the conference, the BGSMath will present examples of successful collaborations with companies such as Seat, BBVA or Horizons Optical.

Pedro Díez, director of a thesis in collaboration with Seat on the quantification of uncertainty in crash simulations, is the director of the Computational Methods and Numerical Analysis (LaCàN) Laboratory of the Polytechnic University of Catalonia, and is also a member of the BGSMath. “We chose a real physical problem presented by industry and we are looking for mathematical tools to come up with a numerical computational solution”, details Díez.

The LaCàN has worked on computational and mathematical modelling for such diverse areas as the simulation of radars in self-driving cars, the propagation of waves to analyse port operativity, or the long-term resistance of concrete containers for nuclear waste.

Contact with industry is “motivating”, according to Díez, “because it helps us to look for relevant problems, and to solve them we showcase our technologies. And it is a challenge because we have to learn to formulate the problem presented by industry in mathematical terms”.

Also the Group of Montse Guillén, director of the Riskcenter at the University of Barcelona, a member of the BGSMath, is working on developing tools to get the most out of data, in this case statistics applied to risk analysis.

“When you work in the real world”, says Guillén, “often you cannot check if the assumptions of your mathematical model are met or not. However, the actual result may not be far from the theoretical one: and this is why specific methods are used”. Also remember that those who work with industry must learn to convey the results in words: “The figures have to be transformed into decisions for companies”, she says.

The Data Science@UB group, led by Jordi Vitrià, is working in the field of machine learning, a branch of artificial intelligence. “Machine learning has a major impact on practical life, even though technically it’s very complicated”, he says. “It is therefore an ideal topic for an industrial PhD”. His group, among others, is working with BBVA with Bayesian models to calculate uncertainty in the field of finance.

Mathematical creativity

According to the director of the BGSMath and professor at the Polytechnic University of Catalonia, Marc Noy, “if industries really wish to make the leap towards digitizing their processes and artificial intelligence, that is to say, industry 4.0, they need the creativity of mathematicians. The BGSMath is already involved in fruitful partnerships with companies, which we wish to strengthen yet further. Industrial PhDs are a perfect tool to provide such cooperation”. “At the same time as we are conducting cutting-edge theoretical and applied research, we wish to train a new generation of mathematics capable of connecting with other disciplines and with other areas outside of academia, such as companies or public authorities. Our ambition is to train highly qualified people and also citizens who are capable of making major changes in society”, he concludes.

Workshop on metaheuristics at UPF

Within the activities of the BGSMath, from 21 to 23 February, the auditorium of the Mercè Rodoreda building on the Ciutadella campus will be hosting a course on metaheuristics taught by recognized experts of this mathematical discipline.

The course, organized jointly by the Business Analytics Research Group of the Department of Economics and Business at UPF and the BGSMath, is intended for master’s degree and PhD students as well as researchers in general who are interested in this subject.