As AI systems move into areas that have real social consequences, trust stops being something you can attribute to an algorithm in isolation. It starts to look more like an outcome of how the whole system behaves over time, including how feedback loops form, how quickly responses arrive, and how institutions react when things go wrong. In this work, we propose a simulation-based approach to study how trust evolves in AI-enabled socio-technical systems. We use Markovian Agent Models (MAMs) to represent users, recommender algorithms, and oversight bodies as interacting populations that change continuously over time. We apply this framework to a misinformation scenario in online recommender systems, with particular attention to situations where engagement-driven optimization allows harmful content to spread before corrective actions take effect. Trust is modeled explicitly as an evaluative index derived from the system’s harmful and corrective states, not as an abstract notion. The simulations suggest that trust erosion is driven mainly by slow or delayed institutional responses, more than by the behavior of individual algorithms on their own. Experiments that remove or delay parts of the governance machinery show that moderation and policy enforcement cannot replace one another: disabling either leads to long-lasting trust loss, and even moderate delays fail to change the system’s trajectory once harmful dynamics have taken hold. The broader point is that trust is shaped by how feedback and timing interact at the system level, and that formal simulation can make these dependencies visible when testing different governance choices under clearly stated assumptions.

Privacy-by-design and privacy-by-default constraints imposed by GDPR require that proper design decisions are taken on systems which deal with personal data in order to avoid legal actions and minimize risk. This impacts architectural decisions and may thus have consequences on system performances. In this paper we present a methodological approach, together with a new version of our specific tool, to support decision processes in privacy-impacted systems and provide assessor-friendly documentation of decision impact. We show the effectiveness of our approach by means of a simple case in health data management systems.

Autonomous Guided Vehicles are one of the hottest research and industrial topics, with applications spanning from alternative transport, security and logistic. However, safety and security is usually one of the main concerns in this type of technology, limiting their application and introduction. In this paper we will consider the effects that a malicious user can cause on the system, and we propose an analysis technique based on a mean-field model, to be able to quickly obtain estimates of the consequences of an attack, and the impact of the counter-measures.

This paper presents a simulation-based methodology to evaluate the performance of a privacy-compliant edge–blockchain architecture for smart city environments. The proposed model combines edge computing with a private, permissioned blockchain to ensure low-latency processing, secure data management, and verifiable transactions. Using a discrete-event simulation framework, we analyze the behavior of the system under realistic workloads and time-varying traffic conditions. The model captures edge operations, including preprocessing and cryptographic tasks, as well as blockchain validation using Proof of Stake consensus. Several experiments explore saturation thresholds, resource utilization, and latency dynamics, under both synthetic and realistic traffic profiles. Results reveal how architectural bottlenecks shift depending on resource allocation and input rate, and demonstrate the importance of balanced dimensioning between edge and blockchain layers.

The GDPR impacts on the design of information systems which process personal data, because it makes mandatory the adoption of the privacy-by-design and privacy-by-default principles. This compliance must be verified along all the design cycle, so that it must be considered as early as possible in the cycle, when alternatives are not yet detailed in the overall design and just general directions of the projects may be available. A comparison between alternatives should be performed, which can only have a qualitative nature, but which involves numerous factors, so a panel of experts is needed to obtained a reliable result. In this paper we propose a AHP-based evaluation approach to examine privacy-related features of alternative information system architectures in the early phases of the design cycle.

This study presents an eXplainable Artificial Intelligence (XAI) framework to predict the performance of marine scrubbers used for sulfur dioxide ( SO 2 ) removal from marine diesel engine flue gases. Using an aggregated dataset from a roll-on/roll-off (Ro-Ro) cargo ship equipped with an open-loop scrubber, combined with satellite data, the study constructs and evaluates multiple artificial intelligence models, including ensemble models, which were benchmarked against each other using standard regression metrics such as the coefficient of determination (R 2 ), mean absolute error (MAE), and mean squared error (MSE). Results achieve high accuracy R 2 > 0 . 92 and offer insights for optimizing scrubber operations. Nevertheless, artificial intelligence models lack transparency. To overcome this problem, this research integrates post-hoc explainability techniques to elucidate the contributions of various features to model predictions, thereby enhancing interpretability and reliability. The integration of SHapley Additive exPlanations (SHAP) and Explain Like I’m 5 (ELI5) not only confirmed the consistency of feature importance rankings (e.g. seawater acidity level, SO 2 inlet concentration, outlet temperature) but also aligned with the physical-chemical principles of SO 2 absorption. Quantitative comparisons with theoretical expectations demonstrated the reliability of the XAI insights, enhancing both model transparency and interpretability. This can improve the current capability of designing scrubber units by defining more efficient and less expensive options for environmental regulation compliance.

Monitoring of non-linear phenomena, such as pollution dynamics, which is the result of several combined factors and the evolution of environmental conditions, greatly benefits by AI tools; a larger benefit derives by the application of explainable solutions, which are capable of providing elements to understand those dynamics for better informed decisions. In this paper we discuss a case with real data in which a posteriori explanations have been produced after the application of ensemble models.

The definition of privacy-related specifications is crucial in the design process of any system which is subject to the GDPR. Privacy-related requirements can be seen as qualitative non-functional requirements which result in additional functional requirements during the specification process. As the application of GDPR is basically assessed by means of risk analysis of data treatments, a quantitative aspect of evaluation is anyway needed: consequently, defining a quantitative approach to privacy-related specifications is desirable, and suitable tools should be identified or provided. While there is some analogy with the field of security and the field of dependability, so that tools might be somehow and to some extent borrowed from those domains, the privacy domain also requires that human factor must be modeled, and external influences on human factors should be modeled as well. In this sense, risk can be evaluated similarly to what can be done in the security field, and this is actually done in the privacy domain by approaches like DPIA, but nudging and deterrence play a different role and are worth some reflections. In this paper we discuss this perspective on privacy-related specification and discuss the use of a tool, Pythia, which is not related to the risk analysis domain but can be profitably used, in our opinion, to define privacy policies as complementary to privacy-aware systems design cycles and to assess their impact. We present an improved analysis of a model from our previous research to show that this point of view on privacy-aware systems design is peculiar and should be considered in the design processes.

Cloud platforms offer not only the capacity to facilitate effective and scalable services for third-party applications and business solutions, but also present an opportunity to implement intricate disaster recovery strategies. For instance, a Chief Technical Officer may opt to maintain operations on private systems in order to effectively manage costs, privacy, and security, while simultaneously leveraging the cloud as an autonomous and immediate disaster recovery support. This can be achieved by building a secondary leg of the IT system that functions as an online cold or hot spare, manages workload peaks, or handles a portion of the workload under normal conditions. To assess the effectiveness and cost-effectiveness of such solutions, appropriate models are essential to examine the trade-offs and explore the parameter space of possible alternatives. This paper presents a modeling approach for the design and evaluation of cloud-based recovery setups.

Modern environmental regulations require rigorous optimization of operations in process engineering to reduce waste, pollution, and risks while maximizing efficiency. However, the nature of chemical plants, which include components with non-linear behavior, challenges the use of consolidated tuning and control techniques. Instead, ad-hoc, self-adapting, and time-variant controls, with a balanced tuning of parameters at both the subsystem and system level, may be necessary. Needed computing processes may require significant resources and high performance systems, if managed by means of traditional approaches and with exact solution methods. In this regard, domain experts suggest instead the use of integrated techniques based on Artificial Intelligence (AI), which include Explainable AI (XAI) and Trustworthy AI (TAI), which are unique in this industry and still in the early stages of development. To pave the way for a real-time, cost-effective solution for this problem, this paper proposes an AI-based approach to model the performance of a real chemical plant, i.e. a marine scrubber installed on a Ro-Ro ship. The study aims to investigate Machine Learning (ML) techniques which can be used to model such processes. Notably, this analysis is the first of its kind, at the best of the authors’ knowledge. Overall, the study highlights the potential of using ML-based techniques, to optimize environmental compliance in the shipping industry.

The diffusion of intelligent services and the push for the integration of computing systems and services in the environment in which they operate require a constant sensing activity and the acquisition of different information from the environment and the users. Health monitoring, domotics, Industry 4.0 and environmental challenges leverage the availability of cost-effective sensing solutions that allow both the creation of knowledge bases and the automatic process of them, be it with algorithmic approaches or artificial intelligence solutions. The foundation of these solutions is given by the Internet of Things (IoT), and the substanding Wireless Sensor Networks (WSN) technology stack. Of course, design approaches are needed that enable defining efficient and effective sensing infrastructures, including energy related aspects. In this paper we present a Domain Specific Language for the design of energy aware WSN IoT solutions, that allows domain experts to define sensor network models that may be then analyzed by simulation-based or analytic techniques to evaluate the effect of task allocation and offloading and energy harvesting and utilization in the network. The language has been designed to leverage the SIMTHESys modeling framework and its multiformalism modeling evaluation features.

Authorship Attribution (AA) is currently applied in several applications, among which fraud detection and anti-plagiarism checks: this task can leverage stylometry and Natural Language Processing techniques. In this work, we explored some strategies to enhance the performance of an AA task for the automatic detection of false and misleading information (e.g., fake news). We set up a text classification model for AA based on stylometry exploiting recurrent deep neural networks and implemented two learning tasks trained on the same collection of fake and real news, comparing their performances: one is based on Federated Learning architecture, the other on a centralized architecture. The goal was to discriminate potential fake information from true ones when the fake news comes from heterogeneous sources, with different styles. Preliminary experiments show that a distributed approach significantly improves recall with respect to the centralized model. As expected, precision was lower in the distributed model. This aspect, coupled with the statistical heterogeneity of data, represents some open issues that will be further investigated in future work.

Modern networking is moving towards exploitation of autonomic features into networks to reduce management effort and compensate the increasing complexity of network infrastructures, e.g. in large computing facilities such the data centers that support cloud services delivery. Autonomicity provides the possibility of reacting to anomalies in network traffic by recognizing them and applying administrator defined reactions without the need for human intervention, obtaining a quicker response and easier adaptation to network dynamics, and letting administrators focus on general system-wide policies, rather than on each component of the infrastructure. The process of defining proper policies may benefit from adopting model-based design cycles, to get an estimation of their effects. In this paper we propose a model-based analysis approach of a simple autonomic router, using Stochastic Petri Nets, to evaluate the behavior of given policies designed to react to traffic workloads. The approach allows a detailed analysis of the dynamics of the policy and is suitable to be used in the preliminary phases of the design cycle for a Software Defined Networks compliant router control plane.

Indoor Object Tracking Systems (IOTS) allow sensing moving objects inside a closed space, where GPS is not available. Besides the most popular use, indoor navigation, IOTS may also contribute to extend the operational range and the possible applications of Virtual Reality (VR) and Augmented Reality (AR) based technologies, such as complex training scenarios or entertainment oriented simulations: in fact, providing devices with a reliable IOTS support adds realism and allows a higher degree of safety and interactivity, that allow a high number of people to take part and collaborate in a simulated scene with a very high degree of physical interaction. In this paper we introduce a novel approach for the optimization and the evaluation of movement tracking in a IOTS based system, oriented to VR/AR applications, with special focus on the training of teams. Our proposal is applied to a case study, an AR application designed to assist business buildings workers in fire extinguisher use training. Performances of our proposal are evaluated by means of a simulation, and results are validated in a test scenario based on Ultra Wide Frequency positioning by means of a simulation scenario fed with real data from anchors.

Microservice-based software architectures are a recent trend, stemming from solutions that have been designed and experimented in big software companies, that aims to support devops and agile development strategies. The main point is that software architectures, similarly to what happens in SOA, are decomposed into very elementary tasks, that can be developed, maintained and deployed in isolation by small independent teams, and that compose an application by means of simple interactions. The resulting architecture is advocated to be more maintainable, less prone to failures, more agile, but obviously impacts on performances. In this paper we provide a simulation based approach to explore the impact of microservicebased software architectures in terms of performances and dependability, given a desired configuration. Our approach aims at giving a first approximation estimation of the behavior of different classes of microservice-based applications over a given system configuration, to characterize the infrastructure from the point of view of the service provider under a randomly generated realistic overall workload: to the best of our knowledge, there is not any other analogous decision support tool available in literature.

The performance over cost ratio of last generation off the shelf devices enables the design of heterogeneous distributed computing systems capable of supporting the implementation of an immersive Virtual Reality, Internet of Things based training support architecture. In this paper we present our work in progress on a low cost distributed immersive simulation system for the training of teams by means of Virtual Reality and off the shelf mobile and prototyping devices. In this case, performance prediction is crucial, because the generation of the scenario have to be performed in real time and synchronization problems may disrupt the result. The approach is demonstrated by a prototypical case study, that consists in a distributed simulator for the interactive training of groups of people that have to coordinate to face a fire emergency, and features advanced immersivity thanks to CGI-enabled stereoscopic 360 degree 3D Virtual Reality and ad-hoc devised interaction interfaces. In particular, we focus on the subsystem that is related to a single trainee, providing a reference implementation and a performance evaluation oriented model to support the design of the complete system.

The Two-hemisphere model-driven (2HMD) approach assumes modelling and use of procedural and conceptual knowledge on an equal and related basis.This differentiates 2HMD approach from pure procedural, pure conceptual, and object oriented approaches.The approach may be applied in the context of modelling of a particular business domain as well as in the context of modelling the knowledge about the domain.Cyber-physical systems are heterogeneous systems, which require multi-disciplinary approach to their modelling.Modelling of cyber-physical systems by 2HMD approach gives an opportunity to transparently compose and analyse system components to be provided and components actually provided, and, thus, to identify and fill the gaps between desirable and actual system content.

Summary Protection and preservation of cultural heritage is an important responsibility for policy makers, public and private institutions, and the citizens themselves. Technologists can make an important contribution by designing monitoring systems for these sites and using the data to prevent incidents. Internet‐of‐things technology offers, for a sustainable price and with significant flexibility, a wide range of different possibilities, fitting different circumstances: from monitoring the environmental parameters of a room in a museum to sensing structural changes in a historical building and to protecting vulnerable artifacts. In this paper, we consider the case of monitoring an extended cultural heritage area: a UNESCO protected site, the center of Matera, an Italian town that will be a European Capital of Culture in 2019. This city is a unique historical settlement, as buildings are partially carved into the rock that constitutes the geological substrate of the area, a local practice used since the prehistoric age. The extent and density of these structures makes the physical protection of the site a big challenge when the expected large crowds of tourists arrive. The objective of the proposed system is to anticipate the threats in a timely manner so that appropriate actions are taken by the authorities thus avoiding damage to the cultural heritage sites. We propose a technique for modeling the performances of the Internet‐of‐things–based monitoring systems that support the planning of incident management in a protected site by exploiting multiple, sparse, heterogeneous, and partially controlled sensors to monitor the behavior of the crowd. The technique is based on the use of Markovian agent models to study the parameters and the dynamics of a scenario, to understand the needs of the monitoring system.

The fashion industry is one of the most flourishing fields for visual applications of IT. Due to the importance of the concept of look in fashion, the most advanced applications of computer graphics and sensing may fruitfully be exploited. The existence of low cost solutions in similar fields, such as the ones that empower the domestic video games market, suggest that analogous low cost solutions are viable and can foster innovation even in small and medium enterprises. In this paper the current state of development of vMannequin, a dynamic, user mimicking, user enacted virtual mannequin software solution, is presented. In order to allow users designing dress concepts, the application simulate the creation and fitting of clothes on virtual models. The interaction is sensor based, in order to both simplify the user interface, and create a richer involvement inside the application.

The effects of networks on the performances of cloud architectures are a very significant issue in designing a data center. The efficiency of data transfers and the overall traffic management are a critical factor that constitutes a potential performance bottleneck, potentially limiting the number of computing nodes that can be installed more than their cost issues. In this paper we present a modeling approach, based on Markovian agents, that allows a performance analysis of network effects in high scale cloud architectures.

Summary The capabilities of multicore processors lead them to be widely adopted in systems at any scale, since their are able to provide more computing power at a lower consumption and dissipation cost. System designers are challenged to a deeper understanding of multicore functioning in order to fully exploit them while keeping the optimal balance between cores utilization and optimal throughput, response time and energy usage. Besides the advancement of general purpose CPUs, the same technological evolution leads to the rise of GPUs, dramatic evolution of graphical coprocessors, that are now affordable, efficient, dedicated computing units, capable of parallel computing and equipped with facilities that make them suited for supporting the main CPU of a system in running ordinary applications. The availability of commercial off‐the‐shelf (COTS) multicore computers, eventually equipped with one or more GPUs, makes them the basic building block of data centers devoted to cloud applications or scientific computing. The way to optimal exploitation of such a wide amount of computing power passes through the ability of matching the best scheduling of hardware resources with the software characteristics of the applications. This requires appropriate models and evaluation methods. Simulation and analytical techniques are essential tools to support the design and the management process of such architectures, but a sound characterization of the workloads is required. Typical workloads consist in multithreaded applications, with different characteristics, that dynamically span over the cores of multiple machines, connected by fast networks. In this paper we propose several parametric performance models for different configurations of multicore machines, with or without GPU support, running multiple class multithreaded applications, aiming to supply a detailed modeling help for complex data centers. Copyright © 2015 John Wiley & Sons, Ltd.

The evaluation of the opportunity of investments on complex production processes is a critical factor in order to enable the balance of risks and potential benefits. There is no out-of-the-box tool that can solve this problem: only the experience of the responsible expert and his knowledgeability of the process can help. Outcome Driven Innovation is an evaluation technique that can support decisions, based on a structured approach to process analysis and on the availability of domain experts: anyway, the need for experts can make the evaluation itself very expensive. In this paper a simulative approach is used to provide an a priori characterization of the conditions that can suggest the opportunity of adopting Outcome Driven Innovation for a process. © ECMS Valeri M. Mladenov, Petia Georgieva, Grisha Spasov, Galidiya Petrova (Editors)

Summary Big Data applications are characterized by a non‐negligible number of complex parallel transactions on a huge amount of data that continuously varies, generally increasing over time. Because of the amount of needed resources, the ideal runtime scenario for these applications is based on complex cloud computing and storage infrastructures, providing a scalable degree of parallelism together with isolation between different applications and resource abstraction. However, such additional abstraction degree also introduces significant complexity in performance modeling and decision making. Potential concurrency of many applications on the same cloud infrastructure has to be evaluated, and, simultaneously, scalability of applications over time has to be studied through proper modeling practices, in order to predict the system behavior as the usage patterns evolve and the load increases. For this purpose, in this paper, we propose an analytic modeling technique based on the use of Markovian Agents and Mean Field Analysis that allows the effective description of different concurrent Big Data applications on a same, multi‐site cloud infrastructure, accounting for mutual interactions, in order to support the careful evaluation of several elements in terms of real costs/risks/benefits for correctly dimensioning and allocating the resources and verifying the existing service level agreements. Copyright © 2014 John Wiley & Sons, Ltd.

Multicore architectures are now available for a wide range of high performance applications, ranging from embedded systems to large scale servers deployed in cloud environments. Multicore architectures are usually subject to two conflicting goals: obtaining a full utilization of the cores while achieving given performance objectives, such as throughput, response time or reduced energy consumption. Moreover, there is a strong interdependence between the software characteristics of the applications, and the underlying CPU architecture. In this scenario, simulation and analytical techniques can provide solid tools to properly design the considered class of systems: however, properly characterize the workload on multithreaded application in multicore environment is not an easy task, and thus is an hot research topic. In this paper we present several models, of increasing complexity, that can characterize multithreaded applications running on multicore architectures.

Big Data applications represent an emerging field, which have proved to be crucial in business intelligence and in massive data management. Big Data promises to be the next big thing in the development of strategical computer applications, even if it requires considerable investment and an accurate resource planning, as the architectures needed to perform at the requisite speed need to scale easily on to a large number of computing nodes. Appropriate management of such architectures benefits from the availability of performance models, to allow developers and administrators to take informed decisions, saving time and experimental work. This paper presents a dedicated modeling language showing firstly how it is possible to ease the modeling process and secondly how the semantic gap between modeling logic and the domain can be reduced.

The design and the requirements of modern computer-based systems have reached a complexity level that calls for the use of models for the verification of non functional requirements since the beginning of their design cycle. Such systems are however too complex to be modeled directly in a simple unstructured formal language like Queueing Networks or Petri Nets. SIMTHESys (Structured Infrastructure for Multiformalism modeling and Testing of Heterogeneous formalisms and Extensions for SYStems) is a novel approach to multiformalism compositional modeling, that is based on the possibility of freely specifying the dynamics of the elements of a formal modeling language in an open framework. This is obtained by the application of consolidated metamodeling foundations to the description of models, together with the concept of behavior as a bridge between formalism dynamics and solution techniques. In this paper the main concepts of the SIMTHESys approach are presented, together with a running example of how SIMTHESys copes with performance evaluation of multiformalism models.

Fault trees are a well known mean for the evaluation of dependability of complex systems. Many extensions have been proposed to the original formalism in order to enhance the advantages of fault tree analysis for the design and assessment of systems. In this paper we propose an extension, repairable fault trees, which allows the designer to evaluate the effects of different repair policies on a repairable system: this extended formalism has been integrated in a multi-formalism multi-solution framework, and it is supported by a solution technique which transparently exploits generalized stochastic Petri nets (GSPN)for modelling the repairing process. The modelling technique and the solution process are illustrated through an example.

The DrawNet++ project addresses the compositional construction of dependability models. Its main goals are to provide: a) a GUI to any graph-based formalism; b) a support to the design process of dependability models, according to concepts inspired by object orientation (OO); c) a user friendly front-end for different classes of analysis/simulation tools.