Airborne particulate matter (PM) poses significant health risks, necessitating accurate toxicity threshold determination for effective risk assessment. This study introduces a novel machine-learning (ML) approach to predict PM toxicity thresholds and identify the key physico-chemical and exposure characteristics. Five machine learning algorithms — logistic regression, support vector classifier , decision tree, random forest, and extreme gradient boosting — were employed to develop predictive models using a comprehensive dataset from existing studies. We developed models using the initial dataset and a class weight approach to address data imbalance. For the imbalanced data, the Random Forest classifier outperformed others with 87% accuracy, 81% recall, and the fewest false negatives (23). In the class weight approach, the Support Vector Classifier minimized false negatives (21), while the Random Forest model achieved superior overall performance with 86% accuracy, 80% recall, and an F1-score of 82%. Furthermore, eXplainable Artificial Intelligence (XAI) techniques, specifically SHAP (SHapley Additive exPlanations) values, were utilized to quantify feature contributions to predictions, offering insights beyond traditional laboratory approaches. This study represents the first application of machine learning for predicting PM toxicity thresholds, providing a robust tool for health risk assessment. The proposed methodology offers a time- and cost-effective alternative to classical laboratory tests, potentially revolutionizing PM toxicity threshold determination in scientific and epidemiological research. This innovative approach has significant implications for shaping regulatory policies and designing targeted interventions to mitigate health risks associated with airborne PM.

In distribution logistics, the planning of vehicles’ routes and vehicles’ loads are traditionally managed separately, despite these activities are correlated. This often leads to various re-designs to make the routes and load plans compatible and applicable in practice. Moreover, the planned routes, which are static by definition, cannot always cope with unexpected events. Traffic congestion, vehicle failures, adverse meteorological conditions, and further undesired events can make the planned routes inapplicable and requirevehicles’ re-routing. This results in lower service levels, undesired delays, and higher costs for logistics companies. With the aim of overcoming the above limitations, this work proposes a novel approach based on a matheuristic algorithm that jointly solves the problem ofdelivery planninganddynamic vehicle routingto automate the delivery process in a logistics 4.0 perspective. The presented algorithm includes two different phases: the static phase, which is executed offline and in advance with respect to the delivery day, and the dynamic phase, which is executed in real-time to cope with unexpected events during the delivery. For the first phase, a matheuristic approach is defined to efficiently solve the combined vehicle routing and loading problems. Differently, for the second phase, a genetic algorithm is proposed to re-route vehicles in real-time, considering both the redefinition in real-time of the nominal trip and/or of the sequence of the customers to be visited. The algorithm is tested both on a literature benchmark and on a real dataset provided by an Italian logistics company. The obtained results show that, on the one hand, the proposed algorithm can automatically provide feasible solutions that minimise travel costs, total travelled distance, and empty space on the vehicles; on the other hand, it can ensure in real-time effective re-routing solutions in case of unexpected events occurring during delivery.Note to Practitioners—This work is motivated by the need for facilitating the operations of planning and routing deliveries in the external logistics sector. We propose an algorithm that automatically generates feasible routing and loading plans for a set of Transport Units (TUs) (i.e., the static phase), and then updates in real-time the nominal route in case of unexpected events (i.e., the dynamic phase). More specifically, the first phase of the algorithm takes as input the set of different clients, the list of products packed into bins (i.e., standard packing units) to be delivered to each client, and the set of transport units available for the deliveries, and provides as output the number and type of TUs to be used, the composition of the bins in each transport unit, and the corresponding route, while optimising the space occupation in each TU and the travel costs. The second phase, instead, takes as input the nominal routes computed in the first phase and, in case of unexpected events (e.g., accidents, slowdowns, etc.) affecting one or more routes, it re-routes the involved trucks guaranteeing the maximum efficiency in regards to travel cost, travel time, and quality of service. The adoption of this algorithm by logistic companies supports the automation of the delivery process and drastically improves the efficiency of logistic operations, with particular regard to the number of used TUs, costs, safety of goods, and customers’ satisfaction.

International audience

In this paper, we propose an innovative meta-heuristic characterized by a multi-dimensional chromosome where each dimension is driven by a rational agent. These agents have to communicate in order to implement evolving and adaptive genetic operators to accelerate the convergence towards the optimal solution. This cooperative approach is applied to solve the patient scheduling problem in emergency department (ED). This problem is NP-difficult due to the permanent interference between three types of arrival: already programmed patients, non-programmed patients and urgent non-programmed patients. Our scheduling problem has to integrate several dimensions such as medical dimensional, patient dimensional, temporal dimensional. The multi-dimensional aspect of the chromosome is crucial to model the different dimensions of the ED. The main goal of the simulation results is to assess the performance of the proposed agent driven multidimensional chromosome. The simulation results confirm that the intra and inter chromosomal interactions allow to avoid the blind aspect of the genetic operators and impacts the quality of solutions. The agents’ cooperation and its ability to improve efficiently the quality of the solutions by exploring intelligently the research space are confirmed by the drop in average total patient waiting time by 15.09%

The early identification of the patients’ hospitalization at triage level within the Emergency Department (ED) presents a potential solution to reduce the risk of overcrowding and improve the quality of care. Thus, predicting patient out-come on arrival assists medical staff in the make of the appropriate patient pathway decision and so reduces the risk of medical error and complication of the patient’s condition. Previous works don’t consider the uncertainty of medical data while the management of this uncertainty is one of the most important and crucial tasks of medical information systems. Thus, we present in this paper an improved version of the classical prediction model by taking into account the uncertainty and by managing properly the missing information. In this context, we propose a new approach based on Dempster-Shafer theory and Dynamic Multi-Layer Perceptron algorithm. Our proposed approach deploys the correspondent neural network as follows: 1) computes for each input parameter the Basic Belief Assignment (BBA) that provides an assessment of the uncertainty pattern using the Dempster’s rule; 2) deduces the correspondent weights based on the computed BBA, and 3) uses an appropriate transfer function to activate the next layer neurons. In this paper, we demonstrate the effectiveness of our proposed method by using a real ED database. We prove that our proposed approach manages efficiently the uncertainty of the medical data sources and missing information, so improves the decision making and reduces errors and complexity.

This paper focuses on scheduling patients in emergency department (ED) according to the priority of patients’ treatments, determined by the triage process. This multi-skills patient scheduling problem is modeled through four dimensional (hypercube) solutions search space whose axes are: Medical staff, Patients, ED structure and Time and it can be formulated as a flexible job shop scheduling problem. We have then to solve a NP-hard combinatorial optimization problem (COP) in the emergency department (ED). The objective is to minimize a score integrating the total waiting time of patients in the (ED) with emphasis on patients with severe conditions. The Friends and Enemies collaboration protocol between agents is developed for solving the problem where each agent integrate a complete metaheuristic scheme in its behavior. Each agent act autonomously in the solution environment and interacts cooperatively with it and with the other agents. The interaction between agents allows the metaheuristic hybridization including the tuning of its parameters. The simulation results show that the scenarios with 2 or more agents were significantly higher in performance than the scenarios with 1 single agent. Thus, it is confirmed that the collaboration protocol between agents influences the quality of the solutions and the scalability of our approach, with the addition of new agents, there is an improvement in the results. Our approach is tested on a set of real (ED) data and the simulation results show that the proposed friends end enemies collaboration protocol can significantly improve the efficiency of the (ED) by reducing the score and especially the total waiting time of multi-skills patient scheduling problem.

In this paper, we study the multi-hop on-demand ridesharing between riders and drivers which are represented as autonomous and rational agents. The goal is to reach the best balance between ridesharing supply and demand. In this context, each agent has its own dynamic perception represented by a bounding box and computed according to its respective constraints and preferences. These perceptions are stored in a spatial R-Tree index allowing users to perform perception overlap queries and identify possible trip shares. The evaluation and selection of optimal path shares is performed by the rider agent based on its objective function. We perform experiments by varying the detour factor of the drivers and demonstrate the validity of our model. We point out the need for optimization on the selection of the optimal transfer node. Finally, we prove the efficiency of our multi-agent based multi-hop ridesharing in terms of service rate and saved distance.

The aim of this paper is to describe our proposed ABOS framework (Agent-Based Optimization Systems) by demonstrating the interest in using the multi-agent approach while operating hybrid metaheuristics to solve Combinatorial Optimization Problems (COP). Two main contributions are highlighted in this work: 1) to show that the alliance of the multi-agent systems (MAS) and the metaheuristics, based on the interaction and the parallelisms concepts, facilitates the hybrid metaheuristics development and allows the simultaneous exploration of different regions of the search space and 2) to demonstrate that the use the multi-agent approach, in the context of optimization, is a crucial option in the process of hybridization allowing the development of generic structures. These later promote the interaction between metaheuristics independent of the problem to be addressed. Our challenge in this ABOS framework is to endow the participant agents, with a set of rational behaviours allowing them to change in real time their strategies, according to the optimization process evolution. The simulation results show that the collaborative optimization can be effective in some cases, hence the need to set effectively the parameters of the optimization algorithms behaviours and the collaborative protocols. We also demonstrate that the use of ABOS framework with MAS allows a more robust and generic structure, capable with minimal changes handling different COP.

In a resource allocation system, resource suppliers and customers can be naturally modelled as autonomous and interactive entities. In this context, we propose in this paper a multiagent system architecture based on trust and honesty concepts between agents in order to synchronize resource allocation in a distributed environment. Indeed, agents use several means in order to allocate resources efficiently as dialogue, adaptability, cooperation, collaboration and even negotiation in addition to the notion of trust. So, individual agents have to evaluate the trustworthiness of others to select those to interact with. Moreover, resource inadequacies can exist in the resource allocation systems, and it is difficult to meet the resource requirements of all agents simultaneously. To overcome these difficulties, we propose a distributed multiagent resource allocation system that emphasizes the issues of agents trust and resource inadequacies, which is called MARA-T&R. In this system, the evidence theory is improved thanks to the Deng entropy to estimate the trustworthiness of agents. Additionally, we use the concept of reservations to solve the problem of resource inadequacies. Our simulation results highlight the excellent performances of this improved trust model and the efficiency of the proposed MARA-T&R system for resource allocation.

People with motor disabilities require assistance for navigating form one location to another. In order to improve the integration of wheelchair users into their daily life and work, we propose a real time adaptive planning algorithm for routing the user through an obstacle free optimal path. Our application is based on an augmented reality system for the assistance of wheelchair people (ARSAWP) and uses augmented reality (AR) smart glasses. The main goal is to support the development of indoor and outdoor navigation systems devoted to wheelchair users. In this paper we detail the design, the implementation and the evaluation of the proposed application, which was implemented in java for the Android operational system. Two types of database are used (local database and remote database). The information about navigation is displayed on AR glasses which give the user the possibility to interact with the system according to the external environment. The prototype is designed for use within the University of Lille campus.

Fuzzy rule-based classification systems (FRBCS) are able to build linguistic interpretable models, they automatically generate fuzzy if-then rules and use them to classify new observations. However, in these supervised learning systems, a high number of predictive attributes leads to an exponential increase of the number of generated rules. Moreover the antecedent conditions of the obtained rules are very large since they contain all the attributes that describe the examples. Therefore the accuracy of these systems as well as their interpretability degraded. To address this problem, we propose to use ensemble methods for FRBCS where the decisions of different classifiers are combined in order to form the final classification model. We are interested in particular in ensemble methods which split the attributes into subgroups and treat each subgroup separately. We propose to regroup attributes by correlation search among the training set elements that belongs to the same class, such an intra-classes correlation search allows to characterize each class separately. Several experiences were carried out on various data. The results show a reduction in the number of rules and of antecedents without altering accuracy, on the contrary classification rates are even improved.

We present an advanced traveler information system (ATIS) for public and private transportation, including vehicle sharing and pooling services. The ATIS uses an agent-based architecture and multi-objective optimization to answer trip planning requests from multiple users in a co-modal setting, considering vehicle preferences and conflicting criteria. At each set of users' requests, the transportation network is represented by a co-modal graph that allows decomposing the trip planning problem into smaller tasks: the shortest routes between the network nodes are determined and then combined to obtain possible itineraries. Using multi-objective optimization, the set of user-vehicle-route combinations according to the users' preferences is determined, ranking all possible route agents' coalitions. The ATIS is tested for the real case study of the Lille metropolitan area (Nord Pas de Calais, France).

Complexity science offers many theories such as chaos theory and coevolutionary theory. These theories illustrate a large set of real life systems and help decipher their nonlinear and unpredictable behaviours. Categorizing an observed Complex System among these theories depends on the aspect that we intend to study, and it can help better understand the phenomena that occur within the system. This article aims to give an overview on Complex Systems and their modelling. Therefore, we compare these theories based on their main behavioural characteristics, e.g. emergence, adaptability, and dynamism. Then we compare the methods used in the literature to model and simulate Complex Systems, and we propose and discuss simple guidelines to help understand one's Complex System and choose the most adequate model to simulate it.

The purpose of this paper is twofold: to depict the patient's journey in operating block of the hospital and to provide a flexible model that matches the reality. In operating block of Jeanne de Flandres Hospital, a daily meeting is held in order to establish the operations that will occur the next day. Organizing a schedule in these circumstances leads to a loss of time and can often lead to conflictual situations. The overall study objective is therefore to provide an interactive schedule. Hence, we will need to consider parameters such as the type of surgery, patient's age, and their past history. From model analysis, some operation's duration will be estimated.

There has been a significant increase in the improvement of response to disasters in crisis management supply chain. Due to their sudden occurrence, these disasters require a consequent quick and efficient response that depends on the ability of logistics systems to generate plans under a variety of constraints. The supply chain studied in this work is a crisis management supply chain, composed of several elements such as transportation means, loading units, suppliers, equipment, resources, persons... We propose an innovative method for solving a distributed delivery scheduling problem, based on a multi-agent system, for the delivery of goods (food, water, clothes, etc.) to the areas affected by the disaster. The covered areas are geographically distributed and partitioned into multiple sub-regions. Each area is assigned a delivery scheduling sub problem. By employing a distributed cooperative framework, we achieved an incorporation of various evaluation parameters in the process of scheduling in order to maintain a high level of synchronization of all the supply chain, and so to insure a better response to the crisis.

The emergent carpooling phenomenon has known a great success throughout the world [1] thanks to efforts developed in this context. Several of the studied works failed and some of them led to operational systems but not without having many drawbacks. Thus, in this paper, we present our works through which we attempt to remedy those limits mainly aiming at providing users with optimized responses to instantaneously issued queries. For this purpose, we primarily focus on useful concepts to reduce real time carpooling's high complexity. The most important among them are a decomposition process to establish a distributed dynamic graph representing the served network and the multi agent concept to perform parallel Optimized and Distributed Assignment of Vehicles (ODAVe) to users' queries.

The co-modality is a new notion introduced by the European commission in 2006. It consists on developing infrastructures and taking measures and actions that will ensure optimum combination of individual transport modes enabling them to be combined effectively in terms of economic, environmental, service and financial efficiency, etc. Including different transport services in one system means that this one must cope with different distributed transport information stored in different location. For these reasons, we propose in this paper a distributed architecture that aims to satisfy transport users by providing them an optimized co-modal itineraries taking in account their constraints and preferences.

In the military, agile and robust military supply chains are key in providing timely responses to support operations in theatre. Supply during peacetime can be managed by proactive logistics plans and classic supply chain management techniques to guaranty the availability of required needs. However, in case of perturbations (time of war, natural disasters...) the need for support increases dramatically and logistics plans need to be adjusted rapidly. To do that, we propose to use anticipation mechanisms. An anticipator uses a world model to construct future predictions about the supply chain and checks whether an undesirable state may be reached. If it is the case, the anticipator adapts his behavior to avoid the actions responsible of this state. In this paper, we focus on the use of anticipation in multi-agent coordination for military logistic planning. We propose de formalize preventive anticipation in the context of a Crisis Management Supply Chain (CMSC).

In this paper, we address the problem of negotiation process in a multi-agents system by using ontologies. Therefore, we present an ontology solution based on the knowledge management system for semantic heterogeneity. The proposed solution prevents the misunderstanding during the negotiation process though the agents' communications. Our approach aims to facilitate, to automate the communications and to make the agents able to understand each other when using these ontologies. Thus, we propose a general architecture for Negotiation process which uses Ontology-based Knowledge Management System (NOKMS). This architecture consists of three layers: the Negotiation Layer (NL) that describes the negotiation process between the Initiator Static Agents (ISAs) and the Participant Mobile Agents (PMAs) by using suitable ontologies, the Semantic Layer (SEL) contains the semantic translator which uses in the case of misunderstanding of the sent messages between the agents, and the last one is the Knowledge Management Systems Layer (KMSL) which bases on the Intelligent Knowledge Base (IKB) to give the flexibility to our negotiation ontology. We will try to apply our architecture on the Multimodal Transport Information System (MTIS) project where we will use an agent architecture in our proposal NOKMS to improves the execution of negotiation process in multi-agents systems which use different ontologies in order to satisfy the transport customers.

This paper explains how to optimize the workplans of mobile agents (MAs) in order to enhance their performance. A MA travels through transport network looking for useful information and services. This information should be available daily to assist travelers to use multimodal transportation system. Our objective is to maximize the number of satisfied transport travelers. In other words, we attempt to increase the satisfaction of transport customers during their travels. So, we intend to optimize task assignment to a set of servers which can offer information with preset cost and processing time. With this intention, we adopted an approach based on evolutionary programs in order to solve our combinatorial problem. Therefore, it is very important to design an efficient representational scheme of a chromosome and develop effective genetic operators. We create a new representation of the chromosome where we integrated the constraints of our problem