ABSTRACT. Myopic bias occurs when early, greedy decisions adversely affect the search trajectory, resulting in obtaining local optima that are globally suboptimal. Aspiration probing derives from a process called Myopic Correction originally proposed with the strategic oscillation approach of tabu search. Aspiration probing is an adaptive memory approach for metaheuristic search that employs periodic probing moves to explore options that are allowed to be reversed if they fail to satisfy the aspiration of yielding an improvement. In this short note, we introduce various simple age-based drop/add approaches for mitigating myopic bias that lay the groundwork for future research in constrained polynomial binary optimization. Testing on a large set of cardinality constrained maximum diversity problems indicates that myopic bias exists and that simple age-based drop/add procedures succeed in reducing it for problems constrained by relatively small cardinality values, pointing the way to advanced forms of aspiration probing to be explored in future work.

ABSTRACT. This paper develops an evolutionary algorithm to solve the Rank Pricing Problem. In this problem, a company establishes prices of a set of products offered to a set of customers in order to maximise its revenue. The proposed algorithm exploits the property that there exists an optimal solution to the problem where prices take values over the set of different customer budgets. As prices can be discretised, pricing decisions become a combinatorial optimisation problem. The algorithm has three distinctive features: the definition of the chromosomes (based on prices), the generation of the initial population, and the application of a local search procedure aiming to improve the current feasible solutions. The computational experience carried out confirms the relevance of the algorithm, especially in terms of the computing time invested.

ABSTRACT. Local search methods start from a feasible solution and improve it by successive minor modifications until a solution that cannot be further improved is encountered. They are a common component of most metaheuristics. Two fundamental local search strategies exist: first-improvement and best-improvement. At any iteration of a first-improvement neighborhood search strategy, the algorithm moves from the current solution to any neighbor with a better (i.e., improvement) value for the objective function. In the case of a best-improvement local search strategy, at any iteration, the algorithm moves from the current solution to its best neighbor whenever the latter improves the former. There is no consensus in the literature regarding which local search strategy performs better regarding solution quality: first-improvement or best-improvement. In this work, we perform an in-depth computational study using rigorous statistical tests in several classes of test problems with different initialization strategies, neighborhood structures, and types of tested instances to evaluate whether one strategy is dominant over the other. Our results reinforce the need for extensive experimentation to decide the most appropriate strategy.

ABSTRACT. This paper introduces the SIRO optimizer, a novel next-generation learned metaheuristic algorithm inspired by biological systems and deep learning techniques. The optimizer uses the susceptible-infected-removed (SIR) epidemiological model to predict the population's susceptibility, active infections, and recoveries. To enhance the search process, SIRO incorporates deep learning into its initialization and parameter tuning components, enabling intelligent and autonomous behaviour. By generating initial solutions based on neural models, the algorithm achieves efficient, effective, and robust search outcomes. To validate the effectiveness of SIRO, a set of numerical test functions from the CEC 2017 benchmark were utilized. The experimental results were compared against various state-of-the-art algorithms, demonstrating that SIRO outperforms its competitors. Moreso, it delivers high-quality solutions while utilizing fewer control parameters. The incorporation of a learning process in SIRO leads to superior precision and computational efficiency compared to other optimization approaches in the existing literature.

ABSTRACT. Multiple Criteria Decision Aiding (MCDA) provides preference models and algorithms to assist decision-makers (DMs) in their decision-making tasks. The preference models are characterized by preference parameters which can be learned through preference learning algorithms from holistic judgments given by the DM. Here, we use Simulated Annealing (SA) to learn the parameters of the Ranking based on Multiple Reference Profiles (RMP) model and its simpler variant SRMP. Extensive experiments demonstrate that our proposal outperforms existing methods in terms of both calculation time and accuracy.

ABSTRACT. With the increasing availability of data and the development of powerful algorithms, deep neural networks have become an essential tool for all sectors. However, it can be challenging to automate the process of building and tuning them, due to the rapid growth of data and their complexity. The demand for handling large amounts of data has led to an increasing number of hidden layers and hyperparameters. A framework or methodology to design the architecture of deep neural networks will be crucial in the future, as it could significantly speed up the process of using deep learning models. We present here a first attempt to create an algorithm that combines aspects of Neural Architecture Search and Hyperparameter Optimization to build and optimize a neural network architecture. The particularity of our algorithm is that it is able to learn how to link neural layers of different types to create increasingly performant neural network architectures. We conducted experiments on four different tasks, including regression, binary and multi-classification, and forecasting, to compare our algorithm with common machine learning models.

ABSTRACT. As digitalization spreads across the globe, the amount of information available is increasing exponentially and users are suffering from information overload. Recommender systems present a feasible and effective means to guide and expose users to products and items which align with their preferences. Specifically with the boom of social networks, collaborative filtering recommender systems offer a means to suggest highly relevant items to a user based on their shared interests with other users in the system. Despite major advancements through the integration of machine learning and hybrid systems, collaborative filtering algorithms struggle to handle large and sparse datasets which hampers the system’s ability to provide accurate recommendations. Metaheuristic techniques have been successful in improving collaborative filtering recommender systems despite data size and sparsity. This study presents a review of different attempts to optimize collaborative filtering recommender systems inclusive of metaheuristic techniques in this evolution which highlights an evident gap in standardized evaluation metrics of recommender systems.

ABSTRACT. The increasing demand for home care services imposes effective human resource management. The problem concerns the creation of teams of one or two caregivers, serving patient requiring one or two caregivers. The number of teams of each type makes up the team scheme. A single-caregiver team can synchronize for tasks requiring two caregivers. Introducing a novel methodology, we employ a biased-randomized greedy constructive algorithm for route design, comparing it with it hybridization with a LS algorithm. Then, the performance of the hybrid method is compared with that of a MILP model and of a BRKGA implementation.

ABSTRACT. Explainability is crucial to building the confidence of the medical team to adopt natural language processing (NLP) techniques. In the majority of recent studies in medical informatics, Deep Learning performed better than machine learning (ML) for natural language processing (NLP) on medical documents. However, the generated models are black-box models difficult to explain. One of these models is word embedding which allows a representation of text and words as vectors, which makes them more exploitable by machines. In this paper, we propose a new method to add explainability to word embedding. We propose a modelization as an optimization problem. The first results on the text8 dataset and 5 target words show the local search can obtain explanations with an improvement of cosine similarity by 11% to 30%.

ABSTRACT. This paper explores the intricate challenges of the elective surgery scheduling problem, considering uncertainties in both surgery duration and length of stay in the intensive care unit. We present a novel two-stage robust approach employing the column-and-constraint generation algorithm to address the master surgical schedule and surgery case assignment problems under these uncertainties. Our approach differs from traditional methods by incorporating a specific modeling of uncertainty using independent uncertainty sets and accounts for surgical teams and resource availability. Comparative analysis with the cutting-plane method demonstrates the effectiveness of our approach, offering valuable insights for the enhanced management of uncertainties in elective surgery planning.

ABSTRACT. The rapid intervention of ambulances when an emergency arises is of vital importance for the well-being and health of the people. The Static Ambulance Location Problem consists of determining the optimal location of emergency vehicles maximizing service coverage and its quality. A novel hybrid methodology is presented, merging Geographic Information Systems and the Greedy heuristic. The methodology is applied to a real-life case in the province of Valencia, Spain. In addition, a graphical tool is used that allows for a global visualization of the impact of different demographic and territorial characteristics. The results show that, thanks to this methodology, it is possible to increase the percentage of the population covered compared to the current situation by optimizing the location of ambulances.

ABSTRACT. In this paper, the continuous and dynamic Berth Allocation Problem is stud-ied to minimize the makespan. A disjunctive graph solution representation is proposed as an alternative to the widely used sequence-based representation. Novel local search algorithms are proposed for the studied problem and em-bedded in an Iterated Local Search metaheuristic. Numerical tests are carried out on randomly generated instances to compare the two solution represen-tation approaches with the results showing that the proposed approach can be successfully deployed to provide good quality solutions for instances of the problem.

ABSTRACT. In this study, we propose a scheduling problem that stems from technological disasters, characterized by risks and the dynamic evolution of hazardous products. The goal is to clean or neutralize these hazardous products to reduce the overall risk of contamination spreading throughout the environment, affecting inhabitants and agricultural areas. The problem is approached as a Resource-Constrained Project Scheduling with Risk and Product Transformation Dynamics. We present a mathematical formulation and introduce an Iterated Local Search metaheuristic. We fine-tune the metaheuristic parameters using a machine learning package and conduct several numerical experiments to assess performance and gain insights into this innovative application.

ABSTRACT. This paper introduces an approach based on the general variable neighborhood search algorithm to address the multi-objective online flexible job shop scheduling problem where new jobs arrive in batches at various times. We focus on minimizing two objectives, the maximum tardiness and the maximum workload. Additional objectives can be easily integrated if necessary. A penalty-based system is proposed to enhance schedule stability and minimize operator fatigue during the rescheduling process whenever a new batch of operations arrives in a workshop. Preliminary results showcase the effectiveness of this approach compared to the offline variant

ABSTRACT. Scheduling of reclaimers activities in dry bulk terminals significantly impact terminal throughput, a crucial performance indicator for such facilities. This study addresses the Reclaimer Scheduling Problem (RSP) while considering periodic preventive maintenance activities for reclaimers. These machines are integral for reclaiming dry bulk materials stored in stockyards, facilitating their loading onto vessels via ship-loaders. The primary aim of the objective function entails the minimization of the overall completion time, commonly referred to as the makespan. Since this problem is NP-hard, we propose a novel greedy constructive heuristic. The solutions obtained from this heuristic serve as the starting point for an efficient General Variable Neighborhood Search (GVNS) algorithm to handle medium-scale instances resembling real stockyard configurations. Computational experiments are conducted by comparing the proposed methods across various problem instances. The results demonstrate that the developed GVNS, coupled with the constructive heuristic for initial solution finding, efficiently improves scheduling efficacy. Thus, it emerges as a new state-of-the-art algorithm for this problem.

ABSTRACT. Maintenance of offshore wind farms (OWF) is a critical task to ensure the reliability of power production, especially considering their higher susceptibility to breakdowns compared to onshore wind farms. In this study, an optimization approach is proposed for the transportation of crew and equipment for OWF maintenance using multiple modes of transportation. The main objective is to minimize the overall operational cost, including fuel cost and crew wages, while ensuring the timely execution of preventive maintenance tasks on OWF. The problem takes into account the use of different types of vessels and a strategy where technicians can be dropped off and continue working independently while the rest of the crew is transported to another OWF terminal. The research is based on the Norwegian case and serves as the basis for the implementation of the Ocean Charge project, which aims to incorporate fully electric vessels for OWF maintenance operations. A mathematical model and a heuristic algorithm are proposed. The preliminary results obtained from the proposed approach are compared with those obtained from an exact method.

ABSTRACT. With the escalating demand for meal delivery services, this study delves into the Meal Delivery Routing Problem (MDRP) within the context of last-mile logistics. Focusing on the critical aspects of courier allocation and order ful-fillment, we introduce a novel approach utilizing a GRASP metaheuristic. The algorithm optimizes the assignment of couriers to orders, considering dynamic factors such as courier availability, order demands, and geograph-ical locations. Real-world instances from a Colombian delivery app form the basis of our computational analysis. Calibration of GRASP parameters re-veals a delicate trade-off between solution quality and computational time. Comparative results with a simulation-optimization based study underscore GRASP's competitive performance, demonstrating strengths in fulfilling or-ders and routing efficiency across diverse instances. This research enhances operational efficiency in the burgeoning food delivery industry, shedding light on practical algorithms for last-mile logistics optimization.

ABSTRACT. Inventory Routing Problems are specially designed to solve transportation problems with inventory management constraints associated. The objective is to serve a set of customers over a finite time horizon, performing product deliveries to meet the demands of customers taking into account constraints regarding inventory levels authorized as well as production capacity. We propose a heuristic method based on an iterative approach that decomposes the original problem into subproblems according to the length of the time horizon. The proposed method is iterative and follows the sequence of periods of the time horizon from the beginning to the end. Therefore, the resolution of a subproblem (with all initial constraints but considering only a part of the time horizon) starts from the resolution of the subproblems for previous periods. The method limits the modification of the solution for periods that has already been considered at an earlier iteration, thus accelerating the resolution. Results shown that our approach is competitive in terms of solution quality and execution time and can provide good solutions for the set of instances considered.