Agent-based Simulation Techniques Research Articles

Agent-based post-earthquake evacuation simulation to enhance early-stage architectural layout and non-structural design

In the indoor design process, architects make crucial decisions regarding architectural layout and the selection of non-structural elements. However, there is a lack of comprehensive consideration for human evacuation behavior, specifically in the event of earthquake evacuation, during the design process. This paper bridges this gap by presenting the application of Agent-Based Building Earthquake Evacuation Simulation (AB2E2S). The paper assesses a post-earthquake evacuation simulation prototype, which integrates an agent-based simulation technique with probabilistic earthquake damage assessment. The model is applied to the case of the engineering building at The University of Auckland, to evaluate the impact of earthquake intensity, design, and behavioral variables on Safe Evacuation Time and number of casualties. Overall, this paper demonstrates the potential of the AB2E2S prototype to inform architects and designers about the effective selection of non-structural elements and architectural layout scenarios for post-earthquake evacuation during the schematic design process.

Influentials, early adopters, or random targets? Optimal seeding strategies under vertical differentiations

Product seeding, defined as the act by which firms send products to selected customers and encourage them to spread word of mouth, is a critical decision support strategy for the success of new products. Using multiple agent-based simulation techniques, we investigated the relative importance of three widely adopted seeding strategies (seeding influentials, early adopters, and random targets) in a competitive market in which products are vertically differentiated in terms of quality and brand strength. We found robust evidence that the finding of an optimal seeding strategy depends on consumers' propensity to spread negative WOM. When negative WOM propensity is low, seeding influentials outperform seeding early adopters or random targets. When negative WOM propensity is high, decision-making about an optimal seeding strategy relies on the relative quality and brand strength of the product and the focal firm's objective. In particular, if a product's relative quality is low, seeding early adopters is the optimal seeding strategy in terms of both market share (MS) and net present value (NPV); if the product's relative quality is equal, seeding early adopters is most effective for increasing MS, while seeding influentials is the best for increasing NPV; and if the product's relative quality is high, seeding influentials is the optimal strategy, except that for products with strong brand strength and firm aims at maximizing the MS growth. We conclude the paper by discussing its theoretical contributions and managerial relevance for decision support.

Advancing autologous CAR T-cell therapy through real-time patient health data integration: a simulation-based approach

Autologous chimeric antigen receptor T-cell therapy presents promising treatment outcomes for various cancers. However, its potential is restrained by unique supply chain challenges, including dynamic patient health conditions and extended turnaround time. These challenges often lead to missed optimal treatment windows, impeding the effective delivery of life-saving treatments. This article presents SimPAC (simulation-based decision support for Patient-centric manufacturing of autologous cell therapies). SimPAC is designed to model and incorporate real-time patient health conditions into the supply chain decisions of autologous chimeric antigen receptor T-cell therapy. SimPAC integrates system dynamics and agent-based simulation techniques, facilitating the adaptation of manufacturing processes and production schedules based on real-time patient health conditions. SimPAC can model various patient disease progressions using parametric functions, nonparametric functions, or tabular data. Additionally, SimPAC offers easy configuration options to model various cell therapy supply chains. We provide two case studies to demonstrate the capabilities of SimPAC and highlight the benefits of patient-centric manufacturing, including improved survival rates and potential economic advantages. However, while the benefits are significant, our study also emphasizes the importance of balancing improved patient outcomes, economic viability and ethical considerations in the context of personalized medicine. SimPAC can be used to explore applications of this approach to diverse therapeutic contexts and supply chain configurations.

Construction Productivity Graph: a comprehensive methodology based on BIM and AI techniques to enhance productivity and safety on construction sites

The construction sector is characterised by distinctive issues, such as product uniqueness, the reluctance to introduce innovation, player fragmentation leading to a low productivity level and a related high level of risk intended due to the increased likelihood of damages and injuries, and the consequences on construction productivity. The common European Union regulatory framework provides strict regulations about on-site working activities, but there is still no standard for the environmental and social sustainability of construction sites. Productivity assumes a crucial role in reducing the environmental impact of construction and positively influences workers’ safety due to higher levels of organisation, reducing time, costs, resource consumption and wasted time. This paper presents a methodology developed by augmenting BIM systems capabilities using Agent-based simulation techniques to simulate and optimise on-site work. The augmented BIM model allows analysis of site conditions in terms of space utilisation and resource allocation, resulting in the ‘Construction Productivity Graph – CPG’. This graphic representation synthesises the results obtained, making it possible to visualise the work progression in different working areas with the resources employed, allowing for the management of productivity rates and the verification of work safety conditions.

Generalised Performance Estimation in Novel Hybrid MPC Architectures: Modeling the CONWIP Flow-Shop System

The ability to supply increasingly individualized market demand in a short period of time while maintaining costs to a bare minimum might be considered a vital factor for industrialized countries’ competitive revival. Despite significant advances in the field of Industry 4.0, there is still an open gap in the literature regarding advanced methodologies for production planning and control. Among different production and control approaches, hybrid architectures are gaining huge interest in the literature. For such architectures to operate at their best, reliable models for performance prediction of the supervised production system are required. In an effort to advance the development of hybrid architecture, this paper develops a model able to predict the performance of the controlled system when it is structured as a controlled work-in-progress (CONWIP) flow-shop with generalized stochastic processing times. To achieve this, we employed a simulation tool using both discrete-event and agent-based simulation techniques, which was then utilized to generate data for training a deep learning neural network. This network was proposed for estimating the throughput of a balanced system, together with a normalization method to generalize the approach. The results showed that the developed estimation tool outperforms the best-known approximated mathematical models while allowing one-shot training of the network. Finally, the paper develops preliminary insights about generalized performance estimation for unbalanced lines.

Modeling Collaborative Behaviors in Energy Ecosystems

The notions of a collaborative virtual power plant ecosystem (CVPP-E) and a cognitive household digital twin (CHDT) have been proposed as contributions to the efficient organization and management of households within renewable energy communities (RECs). CHDTs can be modeled as software agents that are designed to possess some cognitive capabilities, enabling them to make autonomous decisions on behalf of their human owners based on the value system of their physical twin. Due to their cognitive and decision-making capabilities, these agents can exhibit some behavioral attributes, such as engaging in diverse collaborative actions aimed at achieving some common goals. These behavioral attributes can be directed to the promotion of sustainable energy consumption in the ecosystem. Along this line, this work demonstrates various collaborative practices that include: (1) collaborative roles played by the CVPP manager such as (a) opportunity seeking and goal formulation, (b) goal proposition/invitation to form a coalition or virtual organization, and (c) formation and dissolution of coalitions; and (2) collaborative roles played by CHDTs which include (a) acceptance or decline of an invitation based on (i) delegation/non-delegation and (ii) value system compatibility/non-compatibility, and (b) the sharing of common resources. This study adopts a simulation technique that involves the integration of multiple simulation methods such as system dynamics, agent-based, and discrete event simulation techniques in a single simulation environment. The outcome of this study confirms the potential of adding cognitive capabilities to CHDTs and further shows that these agents could exhibit certain collaborative attributes, enabling them to become suitable as rational decision-making agents in households.

Digital twin-enabled automated anomaly detection and bottleneck identification in complex manufacturing systems using a multi-agent approach

Digital twin (DT) models are increasingly being used to improve the performance of complex manufacturing systems. In this context, DTs automatically enabling anomaly detection, such as increase in orders, and bottleneck identification, such as shortage of products, can significantly enhance decision-making to mitigate the consequences of the identified bottlenecks. The existing literature has mainly focused on implementing top-down approaches for analysing the bottlenecks without considering the emergent behaviour of micro-level agents, including inventory levels and human resources, and their impact on the macro-level system’s performance. In order to handle the aforementioned challenges, this paper extends the current literature by proposing a novel DT integrated in a multi-agent cyber physical system (CPS) for detecting anomalies in sensor data, while identifying and removing bottlenecks that emerge during the operation of complex manufacturing systems. An extended 5 C CPS architecture, using multi-agent approach, is implemented to allow DT integration. The agent-based simulation technique enables capturing the probabilistic variability, and aggregate parallelism and dynamism of parallel dynamic interactions within the DT-CPS. A new single agent at the exo-level of the multi-level agent-based modelling structure, called the ‘monitoring agent’, is introduced in this research. The agent detects anomalies and identify bottlenecks through communicating with other agents in different levels automatically. The DT-CPS provides feedback automatically to the physical space to remove and mitigate the identified bottlenecks. The proposed DT based multi-agent CPS has been tested successfully on a real case study in a cryogenic warehouse shop-floor from the cell and gene therapy industry. The performance of the studied cryogenic warehouse is continuously measured using real-time sensor data. The analyses of the results show that the proposed DT-CPS improves the utilisation rates of human resources, on average, by 30% supporting decision making and control in complex manufacturing systems.

Agent-based modelling and simulation for life-cycle airport flight planning and scheduling

ABSTRACT The airport flight planning and scheduling (AFPS) process involves many sub-systems in airport management, including flight arrival prediction, flight landing prioritisation, flight route scheduling, flight departure prioritisation and airport big-data planning. Existing literature has mainly focused on advancing the efficiency of specific sub-system(s) of AFPS. Therefore, there is a call to investigate the AFPS comprehensively and improve its performance systematically, instead of solely enhancing partial sub-system(s) in an airport. This paper proposes a life cycle analysis (LCA) framework to describe the “life cycle” that how an aircraft interacts with the AFPS system of an airport. The agent-based modelling and simulation technique is used to simulate the AFPS system under the LCA framework by treating each aircraft as a passive agent following the orders from the AFPS system. We demonstrate that the naive first-in-first-out principle can be modified to some simple prioritising rules to significantly reduce the overall delay (including importance-weighted delay) and processing time of flights, as well as to increase the number of departed aircraft. Hence, the produced agent-based model allows us to improve AFPS performance by optimising multiple sub-systems simultaneously.

Emergent behaviours in multi-agent systems with Evolutionary Game Theory

The mechanisms of emergence and evolution of collective behaviours in dynamical Multi-Agent Systems (MAS) of multiple interacting agents, with diverse behavioral strategies in co-presence, have been undergoing mathematical study via Evolutionary Game Theory (EGT). Their systematic study also resorts to agent-based modelling and simulation (ABM) techniques, thus enabling the study of aforesaid mechanisms under a variety of conditions, parameters, and alternative virtual games. This paper summarises some main research directions and challenges tackled in our group, using methods from EGT and ABM. These range from the introduction of cognitive and emotional mechanisms into agents’ implementation in an evolving MAS, to the cost-efficient interference for promoting prosocial behaviours in complex networks, to the regulation and governance of AI safety development ecology, and to the equilibrium analysis of random evolutionary multi-player games. This brief aims to sensitize the reader to EGT based issues, results and prospects, which are accruing in importance for the modeling of minds with machines and the engineering of prosocial behaviours in dynamical MAS, with impact on our understanding of the emergence and stability of collective behaviours. In all cases, important open problems in MAS research as viewed or prioritised by the group are described.

A Novel Dynamic Approach for Risk Analysis and Simulation Using Multi-Agents Model

Static risk analysis techniques (SRATs) use event graphs and risk analysis assessment models. Those techniques are not time-based techniques and hence are inadequate to model dynamic stochastic systems. This paper proposes a novel dynamic approach to model such stochastic systems using Dynamic Fault Trees (DFT). The proposed model is called Generic Dynamic Agent-Based Model (GDABM) for risk analysis. GDABM is built on top of the well-known Agent-Based Modeling and Simulation (ABMS) technique. GDABM can model the dynamic system agents in both nominal (failure-free) and degraded (failure) modes. GDABM shows the propagation of failure between system elements and provides complete information about the system’s configurations. In this paper, a complete detailed case study is provided to show the GDABM capabilities to model and study the risk analysis for such dynamic systems. In the case study, the GDABM models the risk analysis for a chemical reactor/operator and performs a complete risk analysis for the entire system. The GDABM managed to simulate the dynamic behavior of the system’s components successfully using Repast Simphony 2.0. Detailed agent behavioral modes and failure modes are provided with various scenarios, including different time stamps. The proposed GDABM is compared to a reference model. The reference model is referred to as the ABM model. GDABM has given very promising results. A comparison study was performed on three performance measures. The performance measures used are (1) Accuracy, (2) response time, and (3) execution time. GDABM has outperformed the reference model by 15% in terms of accuracy and by 27% in terms of response time. GDABM incurs a slightly higher execution time (13%) when compared to the ABM reference model. It can be concluded that GDABM can deliver accepted performance in terms of accuracy and response time without incurring much processing overhead.

How residential energy consumption has changed due to COVID-19 pandemic? An agent-based model.

Integrating occupant behavior with residential energy use for detailed energy quantification has attracted research attention. However, many of the available models fail to capture unseen behavior, especially in unprecedented situations such as COVID-19 lockdowns. In this study, we adopted a hybrid approach consisting of agent-based simulation, machine learning and energy simulation techniques to simulate the urban energy consumption considering the occupants’ behavior. An agent-based model is developed to simulate the in-home and out-of-home activities of individuals. Separate models were developed to recognize physical characteristics of residential dwellings, including heating equipment, source of energy, and thermostat setpoints. The developed modeling framework was implemented as a case study for the Central Okanagan region of British Columbia, where alternative COVID-19 scenarios were tested. The results suggested that during the pandemic, the daily average in-home-activity duration (IHD) increased by approximately 80%, causing the energy consumption to increase by around 29%. After the pandemic, the average daily IHD is expected to be higher by approximately 32% compared with the pre-pandemic situation, which translates to an approximately 12% increase in energy consumption. The results of this study can help us understand the implications of the imposed COVID-19 lockdown with respect to energy usage in residential locations.

Agent-Based Simulation and Modeling of COVID-19 Pandemic: A Bibliometric Analysis

The coronavirus disease has caused an ongoing pandemic worldwide since 2019. To slow the rapid spread of the virus, many countries have adopted lockdown measures. To scientifically determine the most appropriate measures and policies, agent-based simulation and modeling techniques have been employed. It can be challenging for researchers to select the appropriate tools and techniques as well as the input and output parameters. This study conducted a bibliometric analysis, especially a co-word network analysis, to classify relevant research articles into five clusters: conceptual, economic-based, organizational, policy-based, and statistical modeling. It then explained each approach and point of concern. Through this, researchers and modelers can identify the optimal approaches for their agent-based models.

Psychological decision-making process of construction worker safety behavior: an agent-based simulation approach

Objectives. Relationships between safety behavior and its antecedents have been widely studied. However, the psychological decision-making process of construction worker safety behavior (CWSB) is rarely examined from the systematic perspective. Thus, this study constructed the theoretical framework for the decision-making process of CWSB and systematically explored effects of individual factors (education, age and safety knowledge), organizational factors (safety climate and leader–member exchange [LMX]) and psychological factors (psychological capital [PsyCap] and communication competence [CommComp]) on the dynamic performance of CWSB. Methods. Data were collected from the literature and 536 construction workers in China. The theoretical model was tested with the agent-based simulation (ABS) technique. Results. High level of education, safety knowledge, safety climate, LMX, PsyCap and CommComp help to reduce unsafe behavior at the cut-off point. However, the age–safety relationship might present a U-shape, which denotes that reasonable age structure of construction worker groups may be an option for bettering safety performance. The results indicate that the psychological decision-making process of CWSB is not only the result of individual rational decision, but also the product of organizational and psychological impacts. Findings of this study shed lights on safety behavior management practices based on the psychological decision-making process of CWSB.

An agent based simulation system for analyzing stress regulation policies at the workplace

Workplace stress has a significant impact on productivity, since keeping workers’ stress on an adequate level results a key factor for companies to increase their performance. While a high stress level may conduct to anxiety or absenteeism, a low level may also have undesirable consequences, such as lack of motivation. To identify and understand all the elements which interfere on workers’ stress results a key factor in order to improve workers’ performance. However, the complexity of human behavior increases the difficulty of recognizing the influence of these stressors and finding a way to regulate workers’ stress. This paper proposes the use of agent-based simulation techniques for addressing the challenge of analyzing workers’ behavior and stress regulation policies. The main contributions of the paper are: (i) the definition of a stress model that takes into account work and ambient conditions to calculate the stress and the productivity of workers; (ii) the implementation of this model in an agent-based simulation system, enabling the analysis of workplace stress and productivity for different stress regulation policies; (iii) the analysis of four different stress regulation policies; and (iv) the validation of the model with a sensitivity analysis and with its application to a living lab.

An agent-based approach to quantify the uncertainty in Product-Service System contract decisions: A case study in the machine tool industry

Product-service system (PSS) business models appraise the relationship between different stakeholders and focus on a partnership based on profit. Existing literature discusses servitization and the associated cost-benefit analysis (CBA) models mostly from the perspective of original equipment manufacturers. Additionally, CBA is typically conducted using top-down approaches and standard activity-based costing, with limited available data and without considering uncertainty. As a result, inadequate and under-priced contract decisions may be made. To address the problem, this paper extends the current literature by proposing a novel framework for quantifying uncertainty in cost and benefit estimates of PSS contracts. The framework offers a bottom-up costing approach using the agent-based simulation technique. The framework comprises a stochastic CBA model for PSS. It is developed by considering through-life cost and benefit of products and services with aggregate uncertainty in terms of service costs, service lead-times, and their occurrences. The framework has been tested successfully on a real-world case study with a bespoke service provider in the machine tool industry. The model is applied to include spare-parts and availability-based servitization contracts. The simulation results are validated by real-world measurements and expert knowledge. The results involve a comprehensive stochastic analyses of a through-life CBA under probabilistic uncertainty and provide the opportunity to quantify the uncertainty in PSS contract decisions. Moreover, the results highlight that servitization is more beneficial for bespoke service providers in long-term contracts, and for relatively new or retrofitted products. Further research works are required to apply the model on capability-based contracts.

An Agent-Based Model Applied to Brazilian Wind Energy Auctions

This article, for the first time, adopts the agent-based model simulation technique to analyze the pricing process of energy in the Brazilian electricity market (auctions). Within this model, it is possible to analyze how the energy price is affected when a government intervention is observed through the increase in number of public companies participating in the auctions. In this paper, auctions of new and reserve energy of wind power are simulated. Through this model it is possible to compare the choice of bids from participating sellers in the auctions, categorized in two different groups: public and private companies. The agents (sellers) participate in the auctions by learning from the historical and simulated auctions that is regulated by the Brazilian government. Learning is performed through the usage of a variation of the Q-learning algorithm, which provides the sellers the optimal price-bid considering the conditions presented, which means that this price-bid will provide them the maximum reward possible. The results clearly show the average price difference between both generator profiles. In addition, it is possible to state that the price of energy changes due to the relative participation of public or private sellers in the auctions.

Simulating an agent’s decision-making process in black-box managerial environment: An estimation-and-optimisation approach

With the growing need to guide decision-making in today’s complex managerial environment, researchers of the Operations Research/Management Science community have shown a considerable interest in modelling complex managerial systems using the agent-based modelling and simulation technique. This paper presents an estimation-and-optimisation (ESTOPT) architecture to simulate an agent’s decision-making process in black-box managerial environment. An ESTOPT agent’s behaviour is considered as a two-stage process of solving its optimisation problem, some parameters of which are uncertain and need to be estimated. In the first stage, the agent collects and records information for estimation; in the next stage, it attempts to solve the optimisation problem. The solution guides the agent’s actions on the environment which, in turn, provides the agent with new information and payoff as feedback. In this paper, two agent-based models are introduced to demonstrate the implementation of the ESTOPT approach. The simulation outcomes compare favourably with both empirical and theoretical results, suggesting that the ESTOPT approach can be used to simulate an agent’s decision-making process in black-box managerial environment.

Analysis of fraud controls using the PaySim financial simulator

Fraud controls for financial transactions are needed and required by law enforcement agencies to flag suspicious criminal activity. These controls however require deeper analysis of the effectiveness and the negative impact for the legal customers. Owing to the intrinsically private nature of financial transactions this analysis is often performed after several months of actively using fraud controls. In this paper, we present an analysis of different fraud prevention controls on a mobile money service based on thresholds using a simulator called PaySim. PaySim uses aggregated data from a sample dataset to generate a synthetic dataset that resembles the normal operation of transactions and injects malicious behaviour. With technology frameworks such as agent-based simulation techniques, and the application of mathematical statistics, we show in this paper that the simulated data can be as prudent as the original dataset for setting optimal controls for fraud detection.

Ability, Stigma, and Entrepreneurship

This paper sheds light on the relationship among the quality of entrepreneurial entry, the perception of odds of success or failure, and cultural attitudes associated with success or failure by leveraging differing national conditions that govern entrepreneurial choices. We focus on differences in perceived stigma associated with entrepreneurial failure. We set up a theoretical model designed to capture the notion that the population of entrepreneurs in any economy develops through time governed by a hurdle process with feedback. We argue that the height of these hurdles is determined by institutional and cognitive factors. Variation in these factors amongst countries leads to variation in the rate and nature of entrepreneurship and associated innovation. We illustrate the theory using agent-based simulation techniques. We find that the higher the perceived stigma associated with entrepreneurial failure in a nation, the lower average quality of the entrepreneurial pool through adverse selection. Conversel...

Agent-based modeling for the self-management of chronic diseases: An exploratory study

The impact of mobile technologies on healthcare is particularly evident in the case of self-management of chronic diseases, where they can decrease spending and improve life quality of patients. We propose the adoption of agent-based modeling and simulation techniques as built-in tools to dynamically monitor the state of patient health and provide recommendations for self-management. To demonstrate the feasibility of our proposal we focus on Type 1 diabetes mellitus as our case study, and provide simulation results where the dynamic evolution of signal parameters is shown in the case of healthy and Type 1 diabetes mellitus patients, focussing in particular on the beneficial effects that self-management interventions have on plasma glucose values.