Agent-based services for the validation and calibration of multi-agent models

Li, Y.; Brimicombe, Allan J.; Li, C.

doi:10.1016/j.compenvurbsys.2008.09.002

Cited by 11 publications

(5 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consequently, only verification of correct model construction and validation by comparing the results with real data will be discussed further. Verification (also known as internal validation) refers to the correctness of the internal structure, ensuring that the model has been developed in a formally correct manner (e.g., system diagrams, units of measurement, and equations) in accordance with a specified methodology [58]. This implies corroborating that the implemented model conforms to the specifications by running it after changing some initial parameters, as done by other modelers [33].…”

Section: Urban Planner Submodelmentioning

confidence: 96%

“…Several authors have suggested that the validation process should consist of different components: verification, validation, and a sensitivity analysis [12,[53][54][55][56][57][58]. However, since this paper presents a prototype of an ABM, it focuses less on predicting the future and more on understanding and exploring the behavior of the urban system and reproducing specific scenarios.…”

Section: Urban Planner Submodelmentioning

confidence: 99%

See 1 more Smart Citation

Urban Land Allocation Model of Territorial Expansion by Urban Planners and Housing Developers

Cantergiani

Delgado

2017

Environments

View full text Add to dashboard Cite

Agent-based models have recently been proposed as potential tools to support urban planning due to their capacity to simulate complex behaviors. The complexity of the urban development process arises from strong interactions between various components driven by different agents. AMEBA (agent-based model for the evolution of urban areas) is a prototype of an exploratory, spatial, agent-based model that considers the main agents involved in the urban development process (urban planners, developers, and the population). The prototype consists of three submodels (one for each agent) that have been developed independently and present the same structure. However, the first two are based on a land use allocation technique, and the last one, as well as their integration, on an agent-based model approach. This paper describes the conceptualization and performance of the submodels that represent urban planners and developers, who are the agents responsible for officially launching expansion and defining the spatial allocation of urban land. The prototype was tested in the Corredor del Henares (an urban-industrial area in the Region of Madrid, Spain), but is sufficiently flexible to be adapted to other study areas and generate different future urban growth contexts. The results demonstrate that this combination of agents can be used to explore various policy-relevant research questions, including urban system interactions in adverse political and socioeconomic scenarios.

show abstract

Section: Urban Planner Submodelmentioning

confidence: 96%

Section: Urban Planner Submodelmentioning

confidence: 99%

Urban Land Allocation Model of Territorial Expansion by Urban Planners and Housing Developers

Cantergiani

Delgado

2017

Environments

View full text Add to dashboard Cite

show abstract

“…One of the most important aspects in developing a simulation model is its credibility; hence, the validation and verification of any simulation model is necessary. Multi-agent models are difficult to validate because these models represent a new approach to simulation for which traditional validation methods are not always applicable [12]. Given these challenges, it is essential to know what an appropriate validation process is for such models.…”

Section: Verification and Validation Of Mcdaa Simulation Modelmentioning

confidence: 99%

OPNET-Based Performance Analysis of a Multi-agent Architecture for Managing the Mobile Content Delivery Process

Aneiba

Chibelushi

2016

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

This paper addresses the problem of mobile content delivery failure in wireless data networks, and the resulting wastage of communication resources. In these networks, many content delivery transactions fail due to inadequate device or channel capability, possibly after a partial delivery of the content requested by the user. The paper evaluates the performance of a solution devised to enhance and optimise the delivery of mobile content, as a new approach for reducing the probability of wasting valuable communication resources. The proposed solution is a layered multi-agent architecture which is offered in two alternative configurations: a centralised-decision configuration, and a distributed-decision configuration. Furthermore, a baseline configuration (with no agents for managing the content delivery process) is used in the paper, for the purpose of comparative performance evaluation. The simulation results have shown that on average, under heavy traffic conditions and for two levels of device capability (low or high performance device), the distributed-decision configuration outperforms the other two configurations, in terms of lower agent communication overhead, admitting more transactions and reducing bandwidth utilisation. Overall, compared to the baseline system, the layered multi-agent system performs more efficiently in heavy traffic networks and for poor device capability. However, as would be expected, the multi-agent system performs worse than the baseline system under conditions of high device capability, due to the overhead introduced by the communication between agents. The results support the intuitive expectations of agent behaviour in telecommunication systems.

show abstract

“…Agent based modeling (ABM) is a simulation paradigm which has close ties with actual scientific experiments thus making it a good technique for evaluation different paradigms (Li, Brimicombe, & Li, 2008). What makes ABM unique for modeling complex adaptive systems is the way an ABM is typically designed.…”

Section: Agent-based Modelingmentioning

confidence: 99%

A Novel Formal Agent-Based Simulation Modeling Framework of an AIDS Complex Adaptive System

Siddiqa

Niazi

2013

International Journal of Agent Technologies and Systems

View full text Add to dashboard Cite

HIV/AIDS spread depends upon complex patterns of interaction among various sub-sets emerging at population level. This added complexity makes it difficult to study and model AIDS and its dynamics. AIDS is therefore a natural candidate to be modeled using agent-based modeling, a paradigm well-known for modeling Complex Adaptive Systems (CAS). While agent-based models are also well-known to effectively model CAS, often times models can tend to be ambiguous and the use of purely textbased specifications (such as ODD) can make models difficult to be replicated.Previous work has shown how formal specification may be used in conjunction with agent-based modeling to develop models of various CAS. However, to the best of our knowledge, no such model has been developed in conjunction with AIDS. In this paper, we present a Formal Agent-Based Simulation modeling framework (FABS-AIDS) for an AIDS-based CAS. FABS-AIDS employs the use of a formal specification model in conjunction with an agent-based model to reduce ambiguity as well as improve clarity in the model definition. The proposed model demonstrates the effectiveness of using formal specification in conjunction with agent-based simulation for developing models of CAS in general and, social network-based agent-based models, in particular.

show abstract

Agent-based services for the validation and calibration of multi-agent models

Cited by 11 publications

References 35 publications

Urban Land Allocation Model of Territorial Expansion by Urban Planners and Housing Developers

Urban Land Allocation Model of Territorial Expansion by Urban Planners and Housing Developers

OPNET-Based Performance Analysis of a Multi-agent Architecture for Managing the Mobile Content Delivery Process

A Novel Formal Agent-Based Simulation Modeling Framework of an AIDS Complex Adaptive System

Contact Info

Product

Resources

About