Joseangel | José Ángel Martín Baos

A Game-based Model for Pricing and Timetabling in Liberalised Passenger Railway Markets

Sun, 12 Jul 2026 00:00:00 +0000

Abstract

We propose a game-theoretic model for pricing and timetabling in liberalised passenger rail markets under vertical separation. Railway undertakings compete for train paths (on-rail competition) and subsequently set fares facing passenger demand (off-rail competition). Passenger choice is modelled through a nested logit structure. The resulting bi-level equilibrium model determines path bids and ticket prices. A column generation algorithm is developed and validated using data from the Spanish HSR Madrid–Barcelona corridor.

Optimization with constraint learning for pricing services under competition

Sun, 12 Jul 2026 00:00:00 +0000

Abstract

Revenue Management (RM) optimizes pricing and capacity allocation under demand uncertainty. We propose a data-driven RM framework that integrates neural networks or gradient boosting tree models with constraint learning to support joint pricing and capacity decisions. The approach adapts to arbitrary demand patterns learned directly from data, without requiring a predefined demand model, and captures competitive interactions between railway undertakings using only observable information. The methodology is illustrated through a case study of the Madrid-Barcelona high-speed rail corridor.

Code

The code for the experiments presented in this talk is available at under the Apache License 2.0. Please refer to the repository for instructions on how to reproduce the results and access the datasets used in the experiments. Note that the repository contains the current research implementation and the code is still under development, so it may change substantially before publication.

Un enfoque práctico para la enseñanza de Álgebra y Matemática Discreta mediante clase invertida en Ingeniería Informática

Wed, 09 Jul 2025 00:00:00 +0000

Abstract

The teaching of abstract concepts in technical and scientific subjects poses a recurring challenge in university education, particularly in the early years, where students face complex content and new tools. Specifically, in the course on Algebra and Discrete Mathematics, typically taught in the first year of the Computer Engineering degree, these difficulties are compounded by inconsistent study habits and low class participation. Additionally, the content is often perceived as overly abstract and disconnected from real-world applications. This work describes how the practical sessions of this course have been redesigned through the implementation of a flipped classroom methodology. Interactive notebooks in MATLAB, along with explanatory videos available in both Spanish and English, have been developed to promote autonomous study prior to in-person sessions. During class, students collaboratively solve a case study based on a real-world problem, while the instructor acts as a facilitator, guiding discussions and addressing specific questions. The pilot experience, carried out during the 2023/24 academic year, has shown significant improvements in student motivation, understanding, and performance, creating a more participatory and dynamic learning environment.

On 30th September, I received the XXI Abertis Award for sustainable mobility for my doctoral thesis

Mon, 30 Sep 2024 00:00:00 +0000

The Abertis Foundation recognised on 30th September 2024 the best works in the field of transport infrastructure management and road safety in the XXI edition of the Abertis Awards for sustainable mobility. In the doctoral thesis category, the jury awarded the first prize to José Ángel Martín Baos, for his doctoral thesis titled “Machine learning methods applied to transport demand modelling”. This work focuses on the application of advanced machine learning (ML) techniques to optimise transport demand modelling, a crucial aspect for enhancing sustainable mobility and the planning and management of transport infrastructures.

The award, which highlights innovation in transport and road safety research, emphasises the approach taken in this doctoral thesis, demonstrating how machine learning models, such as neural networks, can outperform traditional random utility models (RUM) in predicting transport demand. These models are fundamental in assessing the social and economic impact of new infrastructures and promoting transport policies that drive sustainability.

In my thesis, I also address a significant limitation of ML models: the difficulty in generating reliable econometric indicators, which are essential for infrastructure planning. To address this, I proposed a reinterpretation of a model called Kernel Logistic Regression (KLR), which combines the advantages of RUM with machine learning techniques, achieving more accurate predictions without losing the ability to derive these indicators.

The award ceremony took place at the headquarters of Vocento in Madrid, at the conclusion of the ABC Forum on Sustainable Economy, with the participation of prominent figures in the sector. The awards were presented by Elena Salgado, President of the Abertis Foundation; José Manuel Vassallo, Director of the Abertis-UPM Chair; José Miguel Atienza, Director of the School of Civil Engineering at UPM; and Georgina Flamme, Director of Institutional Relations, Communication and Sustainability of Abertis and the Abertis Foundation.

I shared this recognition with two other outstanding researchers: Tasneem Falah Mohammad Miqdady from the University of Granada, and Javier Caimari Tur from the Polytechnic University of Catalonia. The awardees received a financial grant as well as a sculpture, symbolising the Abertis Foundation’s commitment to research and the development of sustainable mobility solutions.

Photo gallery

An Optimization Model for Determining Access Costs in Rail Freight Transportation

Sun, 30 Jun 2024 00:00:00 +0000

Abstract

The liberalization of European railway markets necessitates a vertical separation governance framework, with Infrastructure Managers (IMs) providing resources and Freight Operating Companies (FOCs) utilizing them. Effective pricing of track access between these entities is vital for profitability and system efficiency. Additionally, achieving a modal shift of freight from road to rail is crucial for road safety and EU objectives alignment. This paper introduces a novel pricing methodology for train paths, optimizing infrastructure utilization by IMs while considering competitive dynamics and environmental impacts. It addresses challenges such as elastic demand, integrating network capacity into pricing strategies, and managing non-additive costs and path interdependencies. The proposed model incorporates dynamic demand patterns, temporally variable capacities, and total revenue considerations across the planning horizon. A discretization-based approach, focusing on irregular time intervals and unit freight tonnage, is developed for solving the continuous model. This involves linking freight tonnage units to a prototype train, with a discrete event simulation model employed. The objective function is also discretized, yielding a finite set of optimization variables to be optimized subject to simulation model. An application of this approach is demonstrated through a case study of the Mediterranean Corridor.

Enhancing the convergence speed of line search methods: Applications in Neural Network training

Sun, 30 Jun 2024 00:00:00 +0000

Abstract

The training of machine learning models, such as neural networks, relies on optimisation techniques that necessitate large volumes of data. The algorithms that have demonstrated satisfactory performance on this task frequently use linear searches on subsets of the data. This ensures that, despite the potential low quality of the search direction, the overall computational cost remains low, making this strategy globally efficient. In these methods, strategies employing a constant learning rate have proven to be particularly effective. This paper introduces a novel scheme designed to significantly expedite the convergence process of line search-based methods. Our approach incorporates additional high-quality linear searches derived from the convergence process of the methods, and by making use of an Armijo rule, it dynamically adjusts the step size through successive reductions or expansions based on the evaluated quality of the descent direction. This strategic adjustment enables more substantial progress in the search direction, potentially reducing the number of iterations needed to reach an optimal solution. We have applied our proposed solution to accelerate the performance of widely-used algorithms such as Gradient Descent (GD), Momentum GD, and Adaptive Moment Estimation (Adam). To illustrate the practical implications and effectiveness of our approach, we present a comprehensive case study focusing on the training of Deep Neural Networks and Kernel Logistic Regression.

Creación de un modelo de probabilidad de victoria para partidos de balonmano mediante técnicas de aprendizaje automático

Wed, 19 Jun 2024 00:00:00 +0000

Abstract

Este trabajo desarrolla y valida un modelo de probabilidad de victoria para partidos de balonmano usando técnicas de aprendizaje automático. A través de un análisis de datos provenientes de federaciones internacionales y europeas de balonmano, se aplicaron métodos como la regresión logística, los árboles de decisión y los random forests para identificar variables críticas que determinan el éxito de un equipo. Se ha realizado una evaluación objetivo de la relevancia de cada indicador dentro de cada uno de los modelos utilizados con el fin de optimizar la precisión de los modelos. Los resultados revelan factores significativos que influyen en la victoria, ofreciendo nuevas perspectivas para mejorar estrategias y rendimientos en balonmano profesional. Además, se sugiere la posibilidad de implementar un sistema de valoración objetiva para jugadores basado en el análisis realizado. Este estudio no solo contribuye al avance en la analítica deportiva aplicada al balonmano sino que también proporciona una herramienta valiosa para entrenadores y aficionados, facilitando la comprensión de los determinantes del éxito en este deporte.

An optimal control model for determining freight rail transport access costs

Sun, 19 May 2024 00:00:00 +0000

Abstract

The liberalization of European railway markets, as outlined in European Directive EU 91/440/EEC, mandates a vertical separation governance structure within the railway industry. This vertical separation involves the establishment of two distinct entities: the Infrastructure Manager (IM), responsible for providing railway resources, and the Freight Operating Company (FOC), which operates freight services utilizing the infrastructure provided by the IM. Determining track access pricing between these entities is of utmost interest for European countries extensive monitorings are carried out has emerged as a significant and urgent challenge, one that has profound implications not only for the profitability of IMs and FOCs but also for the efficiency and quality of the entire railway system.

Furthermore, this issue is pivotal in achieving the target objectives set by the European Union (EU), which aims to achieve a modal shift away from road freight transport towards more sustainable modes of transportation. Specifically, the EU has set a goal of achieving a modal shift of 30% by 2030 and at least 50% by 2050 for shipments exceeding 300 km. In this context, rail freight transport is widely recognized as a potentially cost-effective and environmentally sustainable alternative due to its ability to realize economies of scale, reduce pollutant emissions, and mitigate other externalities.

In this paper, we propose an optimal control method for pricing train paths so that the IM maximizes the utilization of public funding while promoting competition in the rail freight market and taking into account both environmental impact and road safety.

The methodology presents several challenges. First, the model needs to account for elastic demand to capture the effect of costs and travel times on modal split. For this reason, road freight transportation is simplified, and a modal split is performed using a logit model.

Second, it involves replacing a basic toll per kilometer scheme with one that incorporates the capacity of the railway network into the pricing process. This means that congested routes should be subject to higher charges than less-demand routes. Hence, a dynamic cargo flow model with capacities on arcs has been chosen. The temporal aspect of the model allows for consideration of dynamic demand patterns and temporally variable capacities (day/night).

The third element of the model deals with a non-additive cost structure and interdependence of train paths when sharing railway segments. This has been achieved by pricing the total revenue received for the use of train paths over the entire planning period.

On March 4th, 2024, I will give a talk at the NeEDS online seminar series

Tue, 09 Jan 2024 00:00:00 +0000

I am excited to announce that on March 4, at 16:30 CET, I will be taking part in the Online Seminar Series Machine Learning NeEDS Mathematical Optimization, hosted by Emilio Carrizosa, Dolores Romero Morales, and Nuria Gómez Vargas. During this session, I will be presenting “Can machine learning methods effectively model travel mode choice? Beyond predictive performance”. This presentation is the result of a collaborative effort alongside Ricardo García Ródenas, Luis Rodríguez Benítez, Julio Alberto López Gómez, and Tim Hillel. .

About the seminar series

Machine Learning NeEDS Mathematical Optimization is an online seminar series, organized by Emilio Carrizosa (IMUS – Instituto de Matemáticas de la Universidad de Sevilla) and Dolores Romero Morales (CBS – Copenhagen Business School) with the collaboration of PhD Students Nuria Gómez-Vargas (IMUS) and Thomas Halskov (CBS). One of the goals of this activity is to brand the role of Operations Research in Artificial Intelligence with the support of EURO.

This series includes a number of presentations from leading academics in the field of Data Science and Analytics that will cover important topics such as explainability, fairness, fraud, privacy, etc. Mathematical Modeling and Mathematical Optimization will be at the core of their presentations. We also have the YOUNG Online Seminar Series “Machine Learning NeEDS Mathematical Optimization”. In each YOUNG session, three junior academics will show their latest results in this burgeoning area.

The format is a weekly session that takes place every Monday, at 16.30 (CET), and has been active since January 11, 2021. It is 100% online-access, and it has speakers from around the globe. The Online Seminar Series is free thanks to the funding the EU gives to our H2020 MSCA NeEDS project, as well as the support given by Universidad de Sevilla (US) and Copenhagen Business School (CBS). This support is gratefully acknowledged.

ROBIN: Rail mOBIlity simulatioN

Wed, 06 Sep 2023 00:00:00 +0000

Abstract

The use of simulators is frequent in the study of complex systems. They replicate a real system and allow obtaining data from the simulated process as well as providing a mathematical model that helps answer What If? questions. This opens up the possibility of evaluating laboratory environment techniques and processes that will later be implemented in the real system. Competitive passenger rail services are complex systems. The efficient design of market mechanisms that encourage competition to offer more and better services requires analytical tools on which to test these new policies. This makes it essential to develop a simulator, as real and complete as possible, to carry out this analysis. This paper presents the development of a microscopic simulator, called ROBIN (Rail mOBIlity simulatioN), to simulate rail mobility in a competitive regime. The developed simulator is parameterizable and general enough to simulate passenger flows in competing railway systems. The tests carried out with ROBIN using the Spanish railway market as a case of study are proof of its usefulness, allowing disaggregated modelling of passenger behaviour and their travel choices in a competitive railway market.

The time slots allocation problem in liberalised passenger railway markets: a multi-objective approach

Wed, 06 Sep 2023 00:00:00 +0000

Abstract

The liberalisation of the European passenger railway markets through the European Directive EU 91/440/EEC states a new scenario where different Railway Undertakings (RUs) compete with each other in an auction or bidding process for time slots to optimise its profit. The infrastructure resources are provided by the Infrastructure Manager (IM) who analyses and assesses the bids received, allocating the resources to each RU. Time slots allocation is a fact that drastically influences the market equilibrium and a problem whose resolution makes it possible to ensure competitiveness within the market. In this paper, time slots allocation problem is modeled in the context of a liberalised passenger railway market. Then, the problem is addressed using two approaches: the first one allocates time slots to each company according to a set of priorities, while the second one introduces a criterion of fairness in the treatment of companies to ensure competition. A multi-objective algorithm has been developed and evaluated for its effectiveness in solving the time slots allocation problem in a liberalized high-speed corridor of the Spanish railway network.

Nyström-based approximations for kernel logistic regression: Application to transport choice modelling

Mon, 17 Jul 2023 00:00:00 +0000

Abstract

The application of machine learning techniques, more specifically kernel-based techniques, to discrete choice modelling using large datasets is limited by the great number of parameters to be considered when building the kernel matrix and the size of the kernel matrix itself. The spatial and temporal complexity is such that these methods are not applicable to large sample sizes. However, there are techniques that allow generating a low-rank matrix approximation to the kernel matrix, one of them is the Nyström method. One limitation of the Nyström method is that the quality of the kernel matrix approximation depends on the proper choice of landmark points. In this work, four variants of this technique are implemented, a basic uniform method, one based on the K-means algorithm and two different implementations of a non-uniform method based on leverage scores. Later, in the experimentation, we conduct a comparison of these methods applied to two big transport mode choice datasets, which contain a large number of samples and variables. Finally, these results are compared with Multinomial Logit and other techniques currently relevant in the Machine Learning field.

On July 7th, I defended my Ph.D. thesis in Advanced Computer Technologies

Mon, 10 Jul 2023 00:00:00 +0000

On Friday, July 7, 2023, I successfully defended my doctoral thesis titled ‘Machine learning methods applied to transport demand modelling,’ under the supervision of Professors Ricardo García Ródenas and Luis Rodríguez Benítez. The doctoral thesis, which has received the distinction of international mention, has been evaluated by the committee with the highest grade, receiving a unanimous decision of summa cum laude.

Photo gallery

Abstract

Transport demand modelling plays a critical role in transportation planning, enabling the accurate prediction of future transport demand and the evaluation of transport policies and infrastructure plans. However, as transport systems become increasingly complex and recent advances in technology result in massive data collection, traditional analytical methods like Random Utility Models (RUMs) are no longer sufficient to manage this complexity. Therefore, it is necessary to incorporate new techniques to overcome this limitation. This thesis investigates the potential of Machine Learning (ML) methods in this context.

Firstly, it is analysed whether state-of-the-art ML models such as artificial neural networks, support vector machines, and ensemble methods like random forests or gradient boosting decision trees, are superior to RUMs in this research field. To achieve this, the models are compared considering as differential criteria the predictive performance and the ability to derive indicators of decision-makers’ behaviour, always in the context of transport demand modelling.

The results show that classical techniques are outperformed by ML models, but also show that the latter have difficulties in generating reliable econometric indicators. For this reason, a ML model called Kernel Logistic Regression (KLR) is proposed as an alternative to model the utility functions of RUMs, enabling the derivation of econometric indicators. The experiments conducted demonstrate that KLR provides good results on real-world datasets used in previous comparisons in the literature, while providing unbiased estimates of behavioural indicators.

Additionally, it is proposed to extend the application of KLR to a wider range of ML problems by means of an extension of the KLR models called Generalized Kernel Logistic Regression (GKLR). For instance, the GKLR theory has led to the derivation of a novel model called Nested Kernel Logistic Regression (NKLR), which enables the application of KLR to datasets with hierarchically structured data.

Finally, this thesis addresses one of the main limitations of the KLR method, which is the high computational and spatial complexity in large-scale problems. To overcome this limitation, it is suggested the use of the Nyström technique and the implementation of accelerated versions of line search training methods. The results demonstrate that by incorporating these techniques, KLR can efficiently tackle large-scale problems involving hundreds of thousands of data points.

News

Optimization techniques for Kernel Logistic Regression on large-scale datasets: A comparative study

Wed, 28 Jun 2023 00:00:00 +0000

Abstract

In recent years, machine learning techniques have been increasingly applied to modelling the decision-making processes of individuals. One technique that has shown good results in the literature for modelling complex behaviours is the Kernel Logistic Regression (KLR). However, standard KLR implementations have a time complexity of 𝒪(𝑛^{3), which is not feasible for large datasets. To overcome this limitation, one of the purposed alternatives is to approximate the kernel matrix using the Nyström method. The aim of this work is to evaluate the Nyström KLR model on large-scale datasets and to study, at the experimental level, which of the optimisation techniques that allow training this model is the most efficient. As results, the authors show that the Nyström method efficiently computes the objective function and its gradient, enabling the training of KLR models with up to 10}5 parameters. Then, it is evaluated the performance of several optimisation methods, including gradient descend, Momentum, Adam, and L-BFGS-B. It can be concluded that L-BFGS-B is the most efficient method for training the Nyström KLR model. However, given enough computational time and proper hyperparameter tuning, the Adam method can also yield good results.

image title

A Unimodal Ordered Logit model for ranked choices

Sun, 12 Sep 2021 00:00:00 +0000

Abstract

Ordinal scale responses capture qualitative user feedback which can be used to model individual choice preference, or are employed in traffic accident analysis to evaluate accident severity. We present a new choice model for ordinal scale responses in choice tasks that combines a Multinomial Logit model with a Poisson probability mass function. The Poisson distribution, which is suitable for modelling the occurrence of the number of events in a fixed time frame, independent of previous events, can be adapted into the unobserved error distribution of a standard MNL model to capture the natural ordering of the choices by imposing a unimodal constraint on the a posteriori choice probability. In this paper we describe the theoretical framework and the specification of the Unimodal Logit model. We apply our model to evaluate accident severity concerning road collisions. Our results are compared against the traditional ordered logit model and the MNL model.

A comparative study of machine learning, deep neural networks and random utility maximization models for travel mode choice modelling

Wed, 08 Sep 2021 00:00:00 +0000

Abstract

Traditionally, Random Utility Maximization (RUM) models have been widely applied to travel mode choice modelling. Currently, Machine Learning (ML) models are being applied as an alternative to RUM models, since they provide better results in terms of prediction capability and they can manage large volumes of data. In this paper, a comprehensive comparison between classic RUM models and ML models, including single and ensemble classifiers as well as Deep Neural Networks (DNNs), is provided in order to assess systematically the performance of different models over two different datasets which have different sizes and nature of data. Numerical experiments show Random Forest (RF) is the best classifier in terms of accuracy index and the computational cost to train the model.

A Python package for performing penalized maximum likelihood estimation of conditional logit models using Kernel Logistic Regression

Tue, 06 Jul 2021 00:00:00 +0000

DOI:

Abstract

In the last few years, Machine Learning (ML) methods have acquired great popularity due to their success in numerous applications such as autonomous cars, image and voice recognition systems, automatic translation systems, etc. This success has led to an increase in the use of ML methods and the extension of their applications to areas such as transport planning.

One of the main tasks within transport planning is the analysis of transport demand. To do so, it is necessary to analyse the way in which users make their decisions about the trips they make and, therefore, be able to predict the number of passengers on the transport network in relation to respect to interventions made on the transport system. Consequently, transport policies and plans can be evaluated according to the behaviour of the passengers. Discrete choice models based on random utility maximization have been developed over the last four decades and currently they have acquired a high degree of sophistication, becoming the canonical tool for transport demand analysis. Nowadays, the use of ML methods could provide an alternative to discrete choice models, as they offer a high level of accuracy in their predictions. In addition, the analyst is relieved from the need of specifying the functional expressions for the utility functions beforehand.

A Python software package called PyKernelLogit was developed to apply a ML method called Kernel Logistic Regression (KLR) to the problem of predicting the transport demand. This package allows the user to specify a set of models using KLR and the estimation of those using a Penalized Maximum Likelihood Estimation procedure. Moreover, this tool also provides a set of indicators for goodness of fit and the application of model validation techniques. Finally, it allows to obtain the willingness to pay or value of time indicators commonly used in transport planning.

Seminar: Discrete choice modelling using Kernel Logistic Regression

Fri, 18 Jun 2021 09:00:00 +0100

Seminar on the Transport and Mobility Laboratory (Transp-or) on the École polytechnique fédérale de Lausanne (EPFL), Lausanne (Switzerland).

Abstract

During the last years, machine learning methods has gained great popularity due to its success in applications such as autonomous vehicles, intelligent robots, image and voice recognition, etc. This has led to an increased use of these methods and a growing interest in expanding the domain of application of machine learning methods, such as in the field of transport modelling. This seminar presents one of these methods, the Kernel Logistic Regression (KLR), from the point of view of Random Utility Models (RUM). It is presented how KLR can be used to specify the utilities in RUM, freeing the modeler from the need to postulate a functional relation between the features beforehand. A Monte Carlo simulation study is conducted to compare KLR with the Multinomial Logit model, and two of the most promising machine learning methods: the Support Vector Machines and the Random Forests. We have shown that on the simulated data, KLR is the only method that achieves maximum accuracy and leads to an unbiased willingness-to-pay estimator for non-linear phenomena. We have also carried an experiment with a real travel mode choice problem, where Random Forests achieved the highest predictive accuracy, followed by KLR.

Transp-or logo

EPFL logo

A python package for performing penalized maximum likelihood estimation of conditional logit models using kernel logistic regression

Wed, 24 Jun 2020 00:00:00 +0000

Abstract

A webpage to monitor the propagation of COVID-19 disease

Thu, 16 Apr 2020 00:00:00 +0000

The Models and Algorithms for Transport Systems (MAT) group from the University of Castilla-La Mancha (UCLM), to which I belong, has developed a web page for the monitoring, and visualization of the COVID-19 disease. This site, through the use of mathematical techniques, as well as data analysis and artificial intelligence, analyzes the evolution and expansion of the coronavirus with the aim of providing information to monitor the progress of the pandemic. These analytical techniques, according to the group led by UCLM Mathematics Department professor Ricardo Garcia Ródenas, “allow to support decision making processes”.

The website is available at .

For each country, it is possible to view the total number of cases, the active cases, deaths and recoveries. The data for each country is updated daily and has been obtained from a public repository. In addition, for each of the countries, a one-week view forecast of the number of new cases, recoveries and deaths is provided.

Similarly, the website shows the evolution of the virus at the level of the Autonomous Communities in Spain. The data of each one of the communities is daily obtained from the database provided by the Instituto de Salud Carlos III de Madrid, whose data is used by the Spanish Health Ministry. Thus, for each of the Autonomous Communities, the number of total cases, active cases, deaths and recoveries is reported, as well as the number of patients hospitalised or critical (in the ICU). In addition, it is possible to view a one-week view forecast of the number of new cases, recoveries and deaths for each autonomous community.

Finally, the website shows the evolution of the COVID-19 infection rate over time for each country and for all the autonomous communities of Spain, together with a prediction at one week’s time. I is also possible to compare the rate with an equilibrium point that allows to clarify if a country or a certain autonomous community is in an expansion or contraction phase of the epidemic. The aim is to monitor the state of the epidemic and provide information and knowledge to authorities who are working hard on a daily basis to reduce its effects.

Discrete choice modeling using Kernel Logistic Regression

Wed, 18 Sep 2019 00:00:00 +0000

Abstract

The Kernel Logistic Regression is a popular technique in machine learning. In this work this tech- nique is applied to the field of discrete choice modeling. This approach is equivalent to specifying non-parametric utilities in random utility models. A Monte Carlo simulation experiment has been carried out to compare this approach with Multinomial Logit models, comparing the goodness of fit and the capability of obtaining the specified utilities.

A methodology for monitoring traffic flow and air pollution in urban areas

Mon, 29 Jul 2019 00:00:00 +0000

Abstract

Road transportation has become the main source of air pollution in urban areas, which has a major impact on local air quality and human health. For this reason, it is increasingly necessary to accurately estimate the contribution of road transport to air pollution in the cities, so that pollution-reduction measures can be properly designed and implemented appropriately. These pollution reduction measures are becoming increasingly important due to the continued growth in vehicle use and the deterioration of driving conditions (traffic congestion). Authorities find it difficult to meet their environmental objectives and, therefore, reliable mathematical emission models are needed to accurately predict the impact of road transport on air pollution.

Nowadays, intelligent cities are essential to prevent high-level pollution situations and to act when such situations occur. Cities must anticipate pollution peaks and take mitigating measures, such as restricting traffic to a certain number of vehicles or according to their license plates, closing traffic on certain streets, reducing speed limits, etc. In addition, traffic flows must be monitored as they affect pollution levels in that city.

Typical pollution monitoring and control systems often consist of large and expensive devices that are only limited to a few points in the city, hence, they provide information for vast areas and sometimes these systems are not scalable. However, cities are distributed environments where events occur in real time and on a massive scale. Therefore, cities should rely on a low-cost Internet of Things (IoT) infrastructure connected to a cloud platform that supports this type of systems, as well as sensor-based big data applications. These pollution control systems can be combined with a traffic monitoring infrastructure to provide a complete system that can be used as a Decision Support System (DSS) to help authorities make decisions about the environmental impacts caused by pollution before they occur.

In this work, a methodology has been developed for determining the traffic flow and the air pollution in several streets of a city using low-cost devices. An artificial intelligence heuristic algorithm has been employed to process video images from a camera and estimate the traffic flow in the street. This algorithm uses statistical techniques to process the motion vectors generated by the GPU when the video is encoded. Moreover, the computational time required by the algorithm is in the order of 2.5 milliseconds, which allows to count the number of vehicles in real time. This algorithm has been tested proving 90% of accuracy. Finally, this information is uploaded to a cloud service where machine learning techniques can be applied to predict the pollution levels in the city or recommend palliative actions.

Sysorm 2019

Fri, 07 Jun 2019 00:00:00 +0000

This week I have attended the 2nd Spanish Young Statisticians and Operational Researchers Meeting (SYSORM) where I have presented a platform for monitoring air pollution an road traffic in cities.

This meeting has taken place in El Escorial from 5th-7th of June 2019. The meeting, promoted by the Spanish Society of Statistics and Operations Research (SEIO), is organised for and by young researchers (young: with less than three years of postdoctoral experience). The scope of the meeting is to represent and give visibility to the newer generations of talented researchers in statistics and operational research, ranging from methodological to applied research topics, and to foster professional communication between them.

José Ángel Martín talking about a platform for monitoring air pollution and road traffic in cities #sysorm2019 pic.twitter.com/8Lra0As68B
— SYSORM2019 (@sysorm2019) 5 de junio de 2019

Joseangel | José Ángel Martín Baos

A Game-based Model for Pricing and Timetabling in Liberalised Passenger Railway Markets

Abstract

Optimization with constraint learning for pricing services under competition

Abstract

Code

Un enfoque práctico para la enseñanza de Álgebra y Matemática Discreta mediante clase invertida en Ingeniería Informática

Abstract

On 30th September, I received the XXI Abertis Award for sustainable mobility for my doctoral thesis

Related news

Photo gallery

An Optimization Model for Determining Access Costs in Rail Freight Transportation

Abstract

Enhancing the convergence speed of line search methods: Applications in Neural Network training

Abstract

Creación de un modelo de probabilidad de victoria para partidos de balonmano mediante técnicas de aprendizaje automático

Abstract

An optimal control model for determining freight rail transport access costs

Abstract

On March 4th, 2024, I will give a talk at the NeEDS online seminar series

About the seminar series

ROBIN: Rail mOBIlity simulatioN

Abstract

The time slots allocation problem in liberalised passenger railway markets: a multi-objective approach

Abstract

Nyström-based approximations for kernel logistic regression: Application to transport choice modelling

Abstract

On July 7th, I defended my Ph.D. thesis in Advanced Computer Technologies

Photo gallery

Abstract

News

Optimization techniques for Kernel Logistic Regression on large-scale datasets: A comparative study

Abstract

image title

A Unimodal Ordered Logit model for ranked choices

Abstract

A comparative study of machine learning, deep neural networks and random utility maximization models for travel mode choice modelling

Abstract

A Python package for performing penalized maximum likelihood estimation of conditional logit models using Kernel Logistic Regression

Abstract

Seminar: Discrete choice modelling using Kernel Logistic Regression

Abstract

A python package for performing penalized maximum likelihood estimation of conditional logit models using kernel logistic regression

Abstract

A webpage to monitor the propagation of COVID-19 disease

Discrete choice modeling using Kernel Logistic Regression

Abstract

A methodology for monitoring traffic flow and air pollution in urban areas

Abstract

Sysorm 2019