Open Access
A GRASP-based algorithm for solving the emergency room physician scheduling problem
(Elsevier, 2021) Cildoz Esquíroz, Marta; Mallor Giménez, Fermín; Mateo, Pedro; Institute of Smart Cities - ISC
This paper addresses a physician scheduling problem in an Emergency Room (ER) requiring a long-term work calendar to allocate work days and types of shift among all the doctors. The mathematical model is created without simplifications, using the real calendar, including holidays. This precludes the possibility of cyclic-type solutions, and involves numerous and varied constraints (demand, workload, ergonomics, fairness, etc.). An effective solution to this very difficult practical problem cannot be obtained, for large instances, with exact solution methods. We formulate a mathematical representation of a real-world ER physician scheduling problem featuring a hybrid algorithm combining continuous linear programming with a greedy randomized adaptive search procedure (GRASP). Linear programming is used to model a general physician-demand covering problem, where the solution is used to guide the construction phase of the GRASP, to obtain initial full schedules for subsequent improvement by iterative application of Variable Neighborhood Descent Search (VNDS) and Network Flow Optimization (NFO). A computational study shows the superiority of our approach over the Integer Linear Programming method in a set of instances of varying size and difficulty inspired by a real setting. The methodology is embedded in a software tool for generating one-year-ahead physician schedules for a local ER. These solutions, which are now in use, outperform the manually-created schedules used previously. © 2021 Elsevier B.V.
Open Access
Motor unit action potential duration, II: a new automatic measurement method based on the wavelet transform
(Lippincott, Williams & Wilkins, 2007) Rodríguez Carreño, Ignacio; Gila Useros, Luis; Malanda Trigueros, Armando; García Gurtubay, Ignacio; Mallor Giménez, Fermín; Gómez Elvira, Sagrario; Rodríguez Falces, Javier; Navallas Irujo, Javier; Ingeniería Eléctrica y Electrónica; Estadística e Investigación Operativa; Ingeniaritza Elektrikoa eta Elektronikoa; Estatistika eta Ikerketa Operatiboa
To present and evaluate a new algorithm, based on the wavelet transform, for the automatic measurement of motor unit action potential (MUAP) duration. A total of 240 MUAPs were studied. The waveform of each MUAP was wavelet-transformed, and the start and end points were estimated by regarding the maxima and minima points in a particular scale of the wavelet transform. The results of the new method were compared with the gold standard of duration marker positions obtained by manual measurement. The new method was also compared with a conventional algorithm, which we had found to be best in a previous comparative study. To evaluate the new method against manual measurements, the dispersion of automatic and manual duration markers were analyzed in a set of 19 repeatedly recorded MUAPs. The differences between the new algorithm’s marker positions and the gold standard of duration marker positions were smaller than those observed with the conventional method. The dispersion of the new algorithm’s marker positions was slightly less than that of the manual one. Our new method for automatic measurement of MUAP duration is more accurate than other available algorithms and more consistent than manual measurements.
Open Access
Assessing the impact of physicians' behavior variability on performance indicators in emergency departments: an agent-based model
(IEEE, 2025-01-20) Baigorri Iguzquiaguirre, Miguel; Cildoz Esquíroz, Marta; Mallor Giménez, Fermín; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Institute of Smart Cities - ISC
In emergency departments (EDs), traditional simulation models often overlook the variability in physician practice, assuming uniform service provision. Our study introduces a hybrid agent-based discrete-event simulation (AB-DES) model to capture this variability. Through simulation scenarios based on real ED data, we assess the impact of physician behavior on key performance indicators such as patient waiting times and physician stress levels. Results show significant variability in both individual physician performance and average metrics across scenarios. By integrating physician agent modeling, informed by literature from medical and workplace psychology, our approach offers a more nuanced representation of ED dynamics. This model serves as a foundation for future developments towards digital twins, facilitating real-time ED management. Our findings emphasize the importance of considering physician behavior for accurate performance assessment and optimization.
Open Access
Early detection of new pandemic waves: control chart and a new surveillance index
(Public Library of Science, 2024) Cildoz Esquíroz, Marta; Gastón Romeo, Martín; Frías Paredes, Laura; García de Vicuña Bilbao, Daniel; Azcárate Camio, Cristina; Mallor Giménez, Fermín; Institute of Smart Cities - ISC
The COVID-19 pandemic highlights the pressing need for constant surveillance, updating of the response plan in post-peak periods and readiness for the possibility of new waves of the pandemic. A short initial period of steady rise in the number of new cases is sometimes followed by one of exponential growth. Systematic public health surveillance of the pandemic should signal an alert in the event of change in epidemic activity within the community to inform public health policy makers of the need to control a potential outbreak. The goal of this study is to improve infectious disease surveillance by complementing standardized metrics with a new surveillance metric to overcome some of their difficulties in capturing the changing dynamics of the pandemic. At statistically-founded threshold values, the new measure will trigger alert signals giving early warning of the onset of a new pandemic wave. We define a new index, the weighted cumulative incidence index, based on the daily new-case count. We model the infection spread rate at two levels, inside and outside homes, which explains the overdispersion observed in the data. The seasonal component of real data, due to the public surveillance system, is incorporated into the statistical analysis. Probabilistic analysis enables the construction of a Control Chart for monitoring index variability and setting automatic alert thresholds for new pandemic waves. Both the new index and the control chart have been implemented with the aid of a computational tool developed in R, and used daily by the Navarre Government (Spain) for virus propagation surveillance during post-peak periods. Automated monitoring generates daily reports showing the areas whose control charts issue an alert. The new index reacts sooner to data trend changes preluding new pandemic waves, than the standard surveillance index based on the 14-day notification rate of reported COVID-19 cases per 100,000 population.
Open Access
Estimation of patient flow in hospitals using up-to-date data: application to bed demand prediction during pandemic waves
(Public Library of Science, 2023) García de Vicuña Bilbao, Daniel; López-Cheda, Ana; Jácome, María Amalia; Mallor Giménez, Fermín; Institute of Smart Cities - ISC
Hospital bed demand forecast is a first-order concern for public health action to avoid healthcare systems to be overwhelmed. Predictions are usually performed by estimating patients flow, that is, lengths of stay and branching probabilities. In most approaches in the literature, estimations rely on not updated published information or historical data. This may lead to unreliable estimates and biased forecasts during new or non-stationary situations. In this paper, we introduce a flexible adaptive procedure using only near-real-time information. Such method requires handling censored information from patients still in hospital. This approach allows the efficient estimation of the distributions of lengths of stay and probabilities used to represent the patient pathways. This is very relevant at the first stages of a pandemic, when there is much uncertainty and too few patients have completely observed pathways. Furthermore, the performance of the proposed method is assessed in an extensive simulation study in which the patient flow in a hospital during a pandemic wave is modelled. We further discuss the advantages and limitations of the method, as well as potential extensions.
Open Access
Dynamic mean absolute error as new measure for assessing forecasting errors
(Elsevier, 2018-02-14) Frías Paredes, Laura; Mallor Giménez, Fermín; Gastón Romeo, Martín; León, Teresa; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Institute of Smart Cities - ISC
Accurate wind power forecast is essential for grid integration, system planning, and electricity trading in certain electricity markets. Therefore, analyzing prediction errors is a critical task that allows a comparison of prediction models and the selection of the most suitable model. In this work, the temporal error and absolute magnitude error are simultaneously considered to assess the forecast error. The trade-off between both types of errors is computed, analyzed, and interpreted. Moreover, a new index, the dynamic mean absolute error, DMAE, is defined to measure the prediction accuracy. This index accounts for both error components: temporal and absolute. Real cases of wind energy forecasting are used to illustrate the use of the new DMAE index and show the advantages of this new index over other error indices.
Open Access
Design exploration prior to blade multi-disciplinary optimisation
(IOP Publishing, 2018) Echeverría Durá, Fernando; Mallor Giménez, Fermín; San Miguel, Unai; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika
The approach of designing blades as a multi-disciplinary, holistic optimisation implies significant challenges owing to the high complexity of the involved factors such as aerodynamics, elasticity, controller and loads. Moreover, the large number of design variables complicates the intuitive analysis of the relationship between the design variables and responses. This paper presents the design variable exploration prior to blade optimisation, which reveals certain design variable combinations that lead to undesirable dynamic load amplification. Statistical tools, such as multiple logistic regression and fast and frugal decision trees, are applied to identify the conditions causing the phenomenon and predict the possible appearance under new design variable combinations.
Open Access
Including learning and forgetting processes in agent-based simulation models: application to police intervention in out-of-hospital cardiac arrests
(Elsevier, 2025-01-01) Baigorri Iguzquiaguirre, Miguel; Mallor Giménez, Fermín; Estadística, Informática y Matemáticas; Estatistika, Informatika eta Matematika; Institute of Smart Cities - ISC; Universidad Pública de Navarra / Nafarroako Unibertsitate Publikoa
Agent-based modeling has become increasingly popular in recent decades; however, defining agents that accurately depict human behavior remains a significant challenge. This paper contributes to the precise definition of human-like agents by incorporating learning and forgetting processes from the medical and psychological literature into agent-based simulation models. Specifically, the mathematical model for forgetting is developed to be compatible with empirical findings. The empirical evidence also supports the decomposition of the learning process into training sessions and the application of skills in real situations, as followed in this model. The resulting model of learning agents is then applied to study police intervention in out-of-hospital cardiac arrests. In numerous urban areas, there's ongoing discussion regarding the provision of defibrillators in patrol cars and CPR training for police officers. The results demonstrate that including learning and forgetting processes in simulation models provide a more accurate understanding of the benefits of using local police to attend cardiac arrests.
Open Access
Gestión de camas hospitalarias durante la pandemia en Navarra con el apoyo de métodos matemáticos de predicción
(Departamento de Salud del Gobierno de Navarra, 2023) Rodrigo Rincón, Isabel; García de Vicuña Bilbao, Daniel; Esparza Artanga, Laida; Santana-Domínguez, Sergio; Martínez-Larrea, Jesús Alfredo; Mallor Giménez, Fermín; Institute of Smart Cities - ISC
Durante la pandemia por coronavirus, en Navarra se utilizaron modelos matemáticos de predicción para estimar las camas necesarias, convencionales y de críticos, para atender a los pacientes COVID-19. Las seis ondas pandémicas presentaron distinta incidencia en la población, ocasionando variabilidad en los ingresos hospitalarios y en la ocupación hospitalaria. La respuesta a la enfermedad de los pacientes no fue constante en cada onda, por lo que, para la predicción de cada una, se utilizaron los datos correspondientes de esa onda. El método de predicción constó de dos partes: una describió la entrada de pacientes al hospital y la otra su estancia dentro del mismo. El modelo requirió de la alimentación a tiempo real de los datos actualizados. Los resultados de los modelos de predicción fueron posteriormente volcados al sistema de información corporativo tipo Business Intelligence. Esta información fue utilizada para planificar el recurso cama y las necesidades de profesionales asociadas a la atención de estos pacientes en el ámbito hospitalario. En la cuarta onda se realizó un análisis para cuantificar el grado de acierto de los modelos predictivos. Los modelos predijeron adecuadamente el pico, la meseta y el cambio de tendencia, pero sobreestimaron los recursos necesarios para la atención de los pacientes en la parte descendente de la curva. El principal punto fuerte de la sistemática utilizada para la construcción de modelos predictivos fue proporcionar modelos en tiempo real con datos recogidos con precisión por los sistemas de información que consiguieron un grado de acierto aceptable permitiendo una utilización inmediata.
Open Access
Cálculo de la distribución del tiempo de vida de componentes mediante autopsia en sistemas binarios aditivos, serie-paralelo y paralelo-serie
(Universitat Politècnica de Catalunya, 1997) Mallor Giménez, Fermín; Azcárate Camio, Cristina; Pérez Prados, Antonio; Estadística e Investigación Operativa; Estatistika eta Ikerketa Operatiboa
En este artículo se estudia el problema de determinar la función de distribución del tiempo de vida de las componentes de un sistema binario, a partir del conocimiento de las leyes que rigen el funcionamiento del sistema y del conjunto de componentes que causa su fallo (obtenida mediante autopsia del sistema en el momento de su deterioro). Se presentan los resultados de Meilijson (1981) y Nowik (1990) que proponen un sistema de ecuaciones impíıcito para obtener estas distribuciones. Sin embargo, se observa que este sistema es de muy difícil resolución práctica, por lo que nosotros consideramos un método cuya utilización es más restringida pero más sencilla, y estudiamos su aplicación a sistemas binarios aditivos, serie-paralelo y paralelo-serie.