Abstract: E-commerce sales have led to a considerable increase in the demand for last-mile delivery companies, revealing several problems in their logistics processes. Among these problems, are not meeting delivery deadlines. For example, in Chile, the national consumer service (SERNAC) indicated that in 2018, late deliveries represented 23% of complaints in retail online sales and were the second most common reason for complaints. Some of the causes are incorrectly designed delivery zones because in many cases, these delivery zones do not account for the demographic growth of cities. The result is an imbalanced workload between different zones, which leads to some resources being idle while others fail to meet their workload in satisfactory conditions. The present work proposes a hybrid method for designing delivery zones with an objective based on improving the quality of express delivery services. The proposed method combines a preprocess based on the grouping of demand in areas according to the structure of the territory, a heuristic that generates multiple candidates for the distribution zones, and a mathematical model that combines the different distribution zones generated to obtain a final territorial design. To verify the applicability of the proposed method, a case study is considered based on the real situation of a Chilean courier company with low service fulfillment in its express deliveries. The results obtained from the computational experiments show the applicability of the method, highlighting the validity of the aggregation procedure and improvements in the results obtained using the hybrid method compared to the initial heuristic. The final solution improves the ability to meet the conditions associated with express deliveries, compared with the current situation, by 12 percentage points. The results also allow an indicative sample of the critical service factors of a company to be obtained, identifying the effects of possible changes in demand or service conditions

Abstract: The assembly line balancing problem is a classical optimisation problem whose objective is to assign each production task to one of the stations on the assembly line so that the total efficiency of the line is maximized. This study proposes a novel hybrid method to solve the simple version of the problem in which the number of stations is fixed, a problem known as SALBP-2. The hybrid differs from previous approaches by encoding individuals of a genetic algorithm as instances of a modified problem that contains only a subset of the solutions to the original formulation. These individuals are decoded to feasible solutions of the original problem during fitness evaluation in which the resolution of the modified problem is conducted using a dynamic programming based approach that uses new bounds to reduce its state space. Computational experiments show the efficiency of the method as it is able to obtain several new best-known solutions for some of the benchmark instances used in the literature for comparison purposes.

Abstract: A majority automata is a two-state cellular automata, where each cell updates its state according to the most represented state in its neighborhood. A question that naturally arises in the study of these dynamical systems asks whether there exists an efficient algorithm that can be implemented in order to compute the state configuration reached by the system at a given time-step. This problem is called the prediction problem. In this work, we study the prediction problem for a more general setting in which the local functions can be different according to their behavior in tie cases. We define two types of local rules: the stable majority and biased majority. The first one remains invariant in tie cases, and the second one takes the value 1. We call this class the heterogeneous majority cellular automata (HMCA). For this latter class, we show that in one dimension, the prediction problem for HMCA is in NL as a consequence of the dynamics exhibiting a type of bounded change property, while in two or more dimensions, the problem is P-Complete as a consequence of the capability of the system of simulating Boolean circuits.

Abstract: This study aims to analyze and explore criminal recidivism with different modeling strategies: one based on an explanation of the phenomenon and another based on a prediction task. We compared three common statistical approaches for modeling recidivism: the logistic regression model, the Cox regression model, and the cure rate model. The parameters of these models were estimated from a Bayesian point of view. Additionally, for prediction purposes, we compared the Cox proportional model, a random survival forest, and a deep neural network. To conduct this study, we used a real dataset that corresponds to a cohort of individuals which consisted of men convicted of sexual crimes against women in 1973 in England and Wales. The results show that the logistic regression model tends to give more precise estimations of the probabilities of recidivism both globally and with the subgroups considered, but at the expense of running a model for each moment of the time that is of interest. The cure rate model with a relatively simple distribution, such as Weibull, provides acceptable estimations, and these tend to be better with longer follow-up periods. The Cox regression model can provide the most biased estimations with certain subgroups. The prediction results show the deep neural network's superiority compared to the Cox proportional model and the random survival forest.

Abstract: Improving the estimation of origin�destination (O-D) travel times poses a formidable challenge due to the intricate nature of transportation dynamics. Current deep learning models often require an overwhelming amount of data, both in terms of data points and variables, thereby limiting their applicability. Furthermore, there is a scarcity of models capable of predicting travel times with basic trip information such as origin, destination, and starting time. This paper introduces novel models rooted in the k-nearest neighbor (KNN) algorithm to tackle O-D travel time estimation with limited data. These models represent innovative adaptations of weighted KNN techniques, integrating the haversine distance of neighboring trips and incorporating correction factors to mitigate prediction biases, thereby enhancing the accuracy of travel time estimations for a given trip. Moreover, our models incorporate an adaptive heuristic to partition the time of day, identifying time blocks characterized by similar travel-time observations. These time blocks facilitate a more nuanced understanding of traffic patterns, enabling more precise predictions. To validate the effectiveness of our proposed models, extensive testing was conducted utilizing a comprehensive taxi trip dataset sourced from Santiago, Chile. The results demonstrate substantial improvements over existing state-of-the-art models (e.g., MAPE between 35 to 37% compared to 49 to 60% in other methods), underscoring the efficacy of our approach. Additionally, our models unveil previously unrecognized patterns in city traffic across various time blocks, shedding light on the underlying dynamics of urban mobility.

Abstract: The funa is a prevalent concept in Chile that aims to expose a persons bad behavior, punish the aggressor publicly, and warn the community about it. Despite its massive use on the social networks of Chilean society, the real dissemination of funas among communities is unknown. In this paper, we extract, generate, analyze, and compare the Twitter social networks spread of three tweets related to �funas� against three other trending topics, through the analysis of global network characteristics over time (degree distribution, clustering coefficient, hop plot, and betweenness centrality). As observed, funas have a specific behavior, and they disseminate as quickly as a common tweet or more quickly; however, they spread thanks to several network users, generating a cohesive group.

Abstract: Student dropout, defined as the abandonment of a high education program before obtaining the degree without reincorporation, is a problem that affects every higher education institution in the world. This study uses machine learning models over two Chilean universities to predict first-year engineering student dropout over enrolled students, and to analyze the variables that affect the probability of dropout. The results show that instead of combining the datasets into a single dataset, it is better to apply a model per university. Moreover, among the eight machine learning models tested over the datasets, gradient-boosting decision trees reports the best model. Further analyses of the interpretative models show that a higher score in almost any entrance university test decreases the probability of dropout, the most important variable being the mathematical test. One exception is the language test, where a higher score increases the probability of dropout.

Abstract: Constitutional processes are a cornerstone of modern democracies. Whether revolutionary or institutionally organized, they establish the core values of social order and determine the institutional architecture that governs social life. Constitutional processes are themselves evolutionary practices of mutual learning in which actors, regardless of their initial political positions, continuously interact with each other, demonstrating differences and making alliances regarding different topics. In this article, we develop Tree Augmented Naive Bayes (TAN) classifiers to model the behavior of constituent agents. According to the nature of the constituent dynamics, weights are learned by the model from the data using an evolution strategy to obtain a good classification performance. For our analysis, we used the constituent agents' communications on Twitter during the installation period of the Constitutional Convention (July-October 2021). In order to differentiate political positions (left, center, right), we applied the developed algorithm to obtain the scores of 882 ballots cast in the first stage of the convention (4 July to 29 September 2021). Then, we used k-means to identify three clusters containing right-wing, center, and left-wing positions. Experimental results obtained using the three constructed datasets showed that using alternative weight values in the TAN construction procedure, inferred by an evolution strategy, yielded improvements in the classification accuracy measured in the test sets compared to the results of the TAN constructed with conditional mutual information, as well as other Bayesian network classifier construction approaches. Additionally, our results may help us to better understand political behavior in constitutional processes and to improve the accuracy of TAN classifiers applied to social, real-world data.

Abstract: In this article, we investigate the behavior of vehicle platoons operating in a predecessor-following configuration, implemented through sampled-data control systems. Our primary focus is to examine the potential influence of the sampling time on the string stability of the platoon. To address this, we begin by designing a string-stable platoon in continuous time. Subsequently, we consider the controller discretization process and proceed to simulate and implement the designed control strategy on an experimental platform at a scaled-down level. Through experimental testing and some theoretical results, we analyze the effects of different sampling times on the string stability performance of the platoon. We observe that an inappropriate selection of the sampling time can lead to a degradation in string stability within the platoon, making the choice of the sampling time crucial in maintaining the desired string stability properties. These findings highlight the importance of carefully considering the sampling time in the implementation of control systems for platooning applications.