Reinforcement learning generates policies based on reward functions and hyperparameters. Slight changes in these can significantly affect results. The lack of documentation and reproducibility in Reinforcement learning research makes it difficult to replicate once-deduced strategies. While previous research has identified strategies using grounded maneuvers, there is limited work in more complex environments. The agents in this study are simulated similarly to Open Al's hider and seek agents, in addition to a flying mechanism, enhancing their mobility, and expanding their range of possible actions and strategies. This added functionality improves the Hider agents to develop a chasing strategy from approximately 2 million steps to 1.6 million steps and hiders

Modelling and Simulation (M&S) represents one of the fundamental methods to design and study complex systems in many industrial and scientific domains such as transport, energy, and aerospace. M&S techniques enable the analysis and evaluation of many design alternatives while avoiding risks, costs, and failures that come with experimentations on the real system; this opportunity becomes crucial, when realworld tests are too costly to conduct in terms of safety, time, and other resources (Fujimoto et al., 2017). Cloud Computing has captured the interest of the scientific and industrial communities because of the benefits provided by its service models, i.e., Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). These services allow developers to rapidly implement solutions that exploit computing and data storage capacity, network resources, and scalability, without having to deal with common issues related to the configuration of the Cloud infrastructure that is automatically managed by a specific service. Cloud Computing can be defined as a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., servers, storage, and services) that can be rapidly provisioned and released with minimal effort in management (Mell & Grance, 2011). In this context, Cloud, Fog, and Edge computing allow organisations to exploit data processing and storage resources efficiently through the definition of a hierarchical data processing architecture. Fog and Edge computing are both extensions of the Cloud network, where at the lowest level there is the Edge, followed by the Fog level, and finally the Cloud level. The Edge layer allows reducing network traffic as data processing occurs locally on the devices. The Fog layer of the Cloud network architecture pushes intelligence down to the LAN layer, where data are processed in a gateway; thus, the Fog nodes are placed near devices with which it is communicating. Cloud, Fog, and Edge computing can offer suitable services to share and collaborate on M&S projects and perform complex simulation experiments faster and more efficiently through the Modelling and Simulation as a Service (MSaaS) model. While MSaaS provides an everincreasing number of opportunities, it also poses a significant number of pitfalls. One of the major pitfalls is that Cloud infrastructures are massive at all scales; as a consequence, the definition of MSaaS solutions is difficult without a deep knowledge of the involved infrastructures and technologies (Cayirci, 2013). The focus of this special issue is to provide current research results in M&S for Cloud Computing and vice versa. Specifically, the special issue aims at (i) presenting the current state-of-the-art about M&S solutions based on open standards, recent extensions, and innovations related to Cloud Computing technologies; (ii) identifying research directions and technologies that will drive innovations in M&S based on Cloud Computing Infrastructures, and (iii) adopts M&S techniques to formalise and study Cloud Computing environments and services. The application of Cloud-based M&S has increased since 2010 (Mansouri et al., 2020) however, studies, particularly on methodological and technological aspects in the convergence of Cloud Computing and M&S disciplines, continue to be few (Hannay et al., 2021; Tolk, 2020; Zhou et al., 2022). To fill this gap in the literature, we invited researchers and practitioners to submit contributions on conceptual, methodological, and technical advances in High-performance simulation in the Cloud, System Dependability, and Performance Analysis through Big data in the Cloud, Parallel and Distributed simulation through the Cloud services. The resulting contributions deal with a wide range of topics ranging from methodological aspects and development frameworks in adopting M&S in the Cloud Computing environment and vice versa, i.e., how Cloud Computing can support the definition of new M&S methods, models, and techniques. This special issue contains four papers, each of which is briefly described in the following. The paper titled “Mobile Experimentation using Modeling and Simulation in the Fog/Cloud” (by Khaldoon Al-Zoubi and Gabriel Wainer) proposes a method and some algorithms to define Fog nodes as private services in which different middleware run on different Virtual Machines to expose services (AlZoubi & Wainer, 2021). The focus of the research is on the mobility of clients that perform experiments through mobile devices. The proposed method and algorithms take into consideration the position of clients to identify available services that are nearby the Fog zone. To increase mobility assistance, the paper delineates the concept of “mobile simulation experiment”, in which experiments move with the device as it moves away from a specified Fog zone. JOURNAL OF SIMULATION 2022, VOL. 16, NO. 6, 547–549 https://doi.org/10.1080/17477778.2022.2080009

Unintentional lane departure accidents are one of the biggest reasons for the causalities that occur due to human errors. By incorporating lane-keeping features in vehicles, many accidents can be avoided. The lane-keeping system operates by auto-steering the vehicle in order to keep it within the desired lane, despite of changes in road conditions and other interferences. Accurate steering angle prediction is crucial to keep the vehicle within the road boundaries, which is a challenging task. The main difficulty in this regard is to identify the drivable road area on heterogeneous road types varying in color, texture, illumination conditions, and lane marking types. This strenuous problem can be addressed by two approaches, namely, `computer-vision-based approach' and `imitation-learning-based approach'. To the best of our knowledge, at present, there is no such detailed review study covering both the approaches and their related optimization techniques. This comprehensive review attempts to provide a clear picture of both approaches of steering angle prediction in the form of step by step procedures. The taxonomy of steering angle prediction has been presented in the paper for a better comprehension of the problem. We have also discussed open research problems at the end of the paper to help the researchers of this area to discover new research horizons.

Book details Scott E. Page The Model Thinker: What You Need to Know to Make Data Work for You. Hatchett Book Group. 448 pages; ISBN-10: 0465094627; ISBN-13: 978-0465094622. 18.56 USD.

Multi-agent systems (MASs) in the smart-grid area have received a great deal of attention from the research community in recent years. Studies on MAS to the smart grid have brought a number of interesting technical discussions on simulation and modeling of the smart grid and research contributions. Researchers are trying to bring energy efficiency and load balancing in the smart grid. Many of these research works have achieved efficiency in power-system domain, while the social system and consumer satisfaction still need improvement. By focusing on the MAS in smart grid, in this part, we survey the body of knowledge and discuss the challenges of simulation and modeling of MAS in the smart grid. We investigate and group the existing solutions and highlight open-research problems.

Nowadays, one of the key areas of research in smart grid (SG) is demand-response management (DRM). DRM assists in simplifying interactions between the customers and the utility-service providers. It also helps in the improvement of energy efficiency as well as effects on load balancing. Studies on DRM have brought a number of interesting, technical discussions and research contributions. Many of these studies work toward making energy-efficient systems. However, there is a need to work in the domain of customer satisfaction; this area needs considerable new advances. From past few decades, a number of studies have been carried out in SG regarding DRM. However, there is no such work that presents a comprehensive analysis of these works. There is a need to investigate different techniques, their advantages, as well as limitations. By focusing on DRM from a customer satisfaction perspective, in this chapter, we present a detailed overview of different solutions for developing DRM. We also group existing solutions and identify trends and challenges in an SG domain from DRM perspective.

This book chapter discusses two of the agent-based modeling (ABM) levels, i. e. exploratory agent-based modeling (EABM) and validated agent-based modeling (VABM). In first part of this chapter, we shall briefly explain EABM with the help of a case study of 5G networks modeled in an agent-based simulator called NetLogo [1] of the use cases of 5G networks is Internet of Things (IoT). We designed, implemented and experimented this case study to explore the futuristic approaches to ease the implementation of this under-developing 5G networkwhich still needs to be explored. Next, we discuss another important level of modeling, i. e., VABM. Since ABM approach has turned into an attractive and efficient way for displaying large-scale complex systems, verification and validation (V&V) of these models have become questionable. Here, we shall briefly explain VABM with the help of the same case study of 5G networks modeled as in EABM. Using VABM, we shall validate the case study if it is a credible solution.

Book details De Nooy W., Mrvar A., and Batagelj V. Exploratory Social Network Analysis with Pajek 3rd (Expanded) Edition Cambridge: Cambridge University Press; 2018. 334 pages, ISBN 978-1-108-56569-1

Purpose Smart grid can be considered as the next step in the evolution of power systems. It comprises of different entities and objects ranging from smart appliances, smart meters, generators, smart storages, and more. One key problem in modeling smart grid is that while currently there has previously been a considerable focus on the proof of concept aspect of smart grid, there have been very few modeling attempts and even lesser attempts at formalization. To the best of our knowledge, formal specification has not been applied previously in the domain of smart grid. Methods Using a state-based formal specification language namely Z (pronounced as ‘Zed’), we present a novel approach to formally modeling and specify smart grid components. Results The modeling exercise clearly demonstrates that Z is particularly suited for modeling various smart grid components. Conclusions The presented formal specification can be considered as a first step towards the modeling of smart grid using a Software Engineering formalism. It also demonstrates how formal specification can be used to model complex systems in general, and the smart grid, in particular.

Amongst collisions, rear-end collisions are the deadliest. Several rear-end collision avoidance solutions have been proposed recently in the literature. A key problem with existing solutions is their dependence on precise mathematical models. However, real world driving is influenced by a number of nonlinear factors. These include road surface conditions, driver reaction time, pedestrian flow, and vehicle dynamics. These factors involve so many different variations that precise mathematical solutions are hard to obtain, if not impossible. This problem with precise control-based rear-end collision avoidance schemes has also previously been addressed using fuzzy logic, but the excessive number of fuzzy rules straightforwardly prejudices their efficiency. Furthermore, such fuzzy logic-based controllers have been proposed without the use of an appropriate modeling technique. One such modeling technique is agent-based modeling. This technique is suitable because it allows for mimicking the functions of an artificial human driver executing fuzzy rules. Keeping in view these limitations, we propose an enhanced emotion enabled cognitive agent (EEEC_Agent)-based controller. The proposed EEEC_Agent helps autonomous vehicles (AVs) avoid rear-end collisions with fewer rules. One key innovation in its design is to use the human emotion of fear. The resultant agent is very efficient and also uses the Ortony–Clore–Collins (OCC) model. The fear generation mechanism of EEEC_Agent is verified through NetLogo simulation. Furthermore, practical validation of EEEC_Agent functions is performed by using a specially built prototype AV platform. Finally, a qualitative comparison with existing state-of-the-art research works reflects that the proposed model outperforms recent research proposals.

Experimental evaluation of the cooperative multiagent systems (CMASs) provides an assessment way that should be analysed. In this paper, we propose an algorithm with acronym CoopRA that can make a deep performance characterization, based on different indicators, of the experimental evaluation results of a CMAS. This could lead to the formulation of helpful information in some decisions related to the performance of the studied CMASs. In order to validate the proposed algorithm, we performed a case study on a CMAS composed of simple reactive agents that operate by mimicking the problem/task solving of natural ants. We chose this type of cooperative multiagent system architecture, based on the fact that even in case of the cooperative multiagent systems composed of simple efficiently and flexibly cooperating agents could emerges an increased problem solving intelligence at the system's level. The evaluation was performed for the Travelling Salesman Problem (TSP) solving that is a well-known NP-hard problem, having many real-life applications.

Many difficult problems, from the philosophy of computation point of view, could require computing systems that have some kind of intelligence in order to be solved. Recently, we have seen a large number of artificial intelligent systems used in a number of scientific, technical and social domains. Usage of such an approach often has a focus on healthcare. These systems can provide solutions to a very large set of problems such as, but not limited to: elder patient care; medical diagnosis; medical decision support; out-of-hospital emergency care; drug classification among others. A recent key focus is that most of these developed intelligent systems are agent-based approaches, or in other words, they can be considered as agent-based intelligent systems (ABISs). ABISs are formally based on a set of interacting intelligent agents (IAs) in addition to the use of intelligent cooperative approaches namely forming intelligent cooperative multiagent systems (ICMASs). The main direction of study consists in the possibility to measure the artificial systems intelligence, frequently called machine intelligence quotient (MIQ). Recently, we performed some research related to the measuring of the machine intelligence. There is presented a comprehensive review of the scientific literature related to the measuring of the MIQ. We consider that the measuring of the machine intelligence is very actual and important, which could allow the differentiation of ABISs based on their intelligence, choosing of the agent-based systems able to solve the most intelligently specific problems. As the main conclusion of the performed study, we mention that cannot be given a unanimous definition of the ABISs intelligence. Even if the machine intelligence cannot be defined, it could be measured. We discuss this affirmation more in-depth in the paper. This is similar to the human intelligence that is not understood very well but can be measured using human intelligence tests.

A mathematician who can only generalise is like a monkey who can only climb up a tree, and a mathematician who can only specialise is like a monkey who can only climb down a tree.In fact neither the up monkey nor the down monkey is a viable creature.A real monkey must find food and escape his enemies and so must be able to incessantly climb up and down.A real mathematician must be able to generalise and specialise."George Polya quoted in D MacHale, Comic Sections (MacHale 1993).Seventh of September marks the death anniversary of the famous mathematician George Polya.Polya is famed for his specific problem-solving techniques given in the bestseller "How to Solve it" (Polya 2004).Polya rightly notes the importance of abstraction.Complex adaptive systems involve the interaction of numerous components or agents.To be able to effectively model these systems, it is extremely important to use abstraction.Abstraction, however, is not limited to a particular domain.We use abstraction in our everyday lives and in social interactions.The sciences are filled with examples of abstraction-double-helix/staircase model for the DNA molecule (Watson 2012), Maxwell's demon in Physics (Maxwell 1891) or rules in Medical diagnosis systems (Iantovics 2012; Iantovics et al. 2018).Not lacking behind are the humanities with concepts such as truth, liberty, or freedom among others.Perhaps the importance of abstraction can be highlighted by the fact that the way humans learn is by means of mechanisms entirely based on abstraction-e.g. a child once affected by something "hot" can later be guided to stay away from other dangerous items by means of a suitable simile.Perhaps being able to easily abstract and frequently use abstraction in our cognitive and thought processes is what separates us from the machines that we design-this being true, at least to-date.

Clinical Decision Support Systems (CDSS) form an important area of research. In spite of its importance, it is difficult for researchers to evaluate the domain primarily because of a considerable spread of relevant literature in interdisciplinary domains. Previous surveys of CDSS have examined the domain from the perspective of individual disciplines. However, to the best of our knowledge, no visual scientometric survey of CDSS has previously been conducted which provides a broader spectrum of the domain with a horizon covering multiple disciplines. While traditional systematic literature surveys focus on analyzing literature using arbitrary results, visual surveys allow for the analysis of domains by using complex network-based analytical models. In this paper, we present a detailed visual survey of CDSS literature using important papers selected from highly cited sources in the Thomson Reuters web of science. We analyze the entire set of relevant literature indexed in the Web of Science database. Our key results include the discovery of the articles which have served as key turning points in literature. Additionally, we have identified highly cited authors and the key country of origin of top publications. We also present the Universities with the strongest citation bursts. Finally, our network analysis has also identified the key journals and subject categories both in terms of centrality and frequency. It is our belief that this paper will thus serve as an important role for researchers as well as clinical practitioners interested in identifying key literature and resources in the domain of clinical decision support.

Online social media has completely transformed how we communicate with each other. While online discussion platforms are available in the form of applications and websites, an emergent outcome of this transformation is the phenomenon of ‘opinion leaders’. A number of previous studies have been presented to identify opinion leaders in online discussion networks. In particular, Feng (2016 Comput. Hum. Behav. 54, 43–53. (doi:10.1016/j.chb.2015.07.052)) has identified five different types of central users besides outlining their communication patterns in an online communication network. However, the presented work focuses on a limited time span. The question remains as to whether similar communication patterns exist that will stand the test of time over longer periods. Here, we present a critical analysis of the Feng framework both for short-term as well as for longer periods. Additionally, for validation, we take another case study presented by Udanor et al. (2016 Program 50, 481–507. (doi:10.1108/PROG-02-2016-0011)) to further understand these dynamics. Results indicate that not all Feng-based central users may be identifiable in the longer term. Conversation starter and influencers were noted as opinion leaders in the network. These users play an important role as information sources in long-term discussions. Whereas network builder and active engager help in connecting otherwise sparse communities. Furthermore, we discuss the changing positions of opinion leaders and their power to keep isolates interested in an online discussion network.

The use of online social media is also connected with the real world. A very common example of this is the effect of social media coverage on the chances of success of elections. Previous literature has identified that the outcome of elections can often be predicted based on online public discussions. These discussions can be across various online social network with a special focus on the candidate's own accounts. Among many other forms of social media, Wikipedia is a very widely-used self-organizing information resource. The management and administration of Wikipedia is performed using special users which are elected by means of online public elections. In other words, the results of these elections pose as an emergent outcome of a large-scale self-organized opinion formation process. However, due to dynamical, and non-linear interactions besides the presence of mutual dependencies between election participants, a statistical analysis of this data can both be cumbersome as well as inefficient in terms of information extraction. We believe that social network analysis is a more appropriate alternative. It allows for the identification of local and global patterns, identification of influential nodes as well as the contacts involved in the influence. In general, this particular analytic technique can help in examining the internal complex network dynamics. In the current paper, we investigates whether personal contacts matter more than know-how contacts in wiki election nominations and voting participation. We employ the use of standard social network analysis tools such as Pajek and Gephi. The presented work demonstrates the significance of personal contacts over know-how contacts of a person in online elections. We have discovered that personal contacts, i.e. immediate neighbors (based on degree centrality) and neighborhood (k-neighbors) of a person have a positive effect on a person’s nomination as an administrator and also contribute to the active participation of voters in voting. Moreover, know-how contacts, analyzed by means of measures such as betweenness and closeness centralities, have a relatively insignificant effect on the selection of a person. However, know-how contacts, measured in terms of betweenness centrality can positively contribute only to the voting process—primarily due to the role played in passing information around the network. These contacts, also measured in terms of influence domain and PageRank, can play a vital role in the selection of an admin. Additionally, such contacts have a positive association with the voting process in terms of reachability and brokerage roles.