Search results

The purpose of this paper is to present research in the area of the modeling of complex systems using feed‐forward neural network.

Applications of multilayer neural networks with supervisor learning on the own simulator program wrote in Borland^® Pascal Language. Series‐parallel identification method is applied. Tapped delay lines (TDL) in static neural networks for modeling of dynamic plants are used. Gradient and heuristic learning algorithms are applied. Three kinds of calibration of learning and testing data are used.

This paper illustrates that feed‐forward multilayer neural networks can model complex systems. Feed‐forward multilayer neural networks with TDL can be used to build global dynamic models of complex systems. It is possible to compare the quality both models.

The learning and testing data from real systems to tune neuronal models require use of calibrating these data to range 0‐1.

The models quality depends on kind of calibration learning data from real system and depends on kind of learning algorithms.

The method and the learning algorithms discussed in the paper can be used to create global models of complex systems. The multilayer neural network with TDL can be used to model complex dynamic systems with low dynamics.

Complex system modeling requires not only understanding of modeling framework but also domain knowledge of the system. The purpose of this paper is to present an approach which separates the domain knowledge from the modeling framework with different views.

By establishing the mechanism of association and fusion among the views, the description and characterization of system from different aspect and point of view can form a complete system model. Based on the approach, a modeling and simulation (M&S) platform named SimFaster is developed. Modeling environment and simulation engine are the most important parts of the platform. The modeling environment provides multi‐views and multi‐layers to help the developers to modeling the structure, layers, composition, behavior, and interactions of an application system. The simulation engine provides mechanism of integration and interaction for components and objects, and provides runtime support for the concepts and terms from modeling environment. The simulation engine organizes the objects in the memory of distributed system as reflective object database system, so it is repository centered architecturally.

Based on the approach of multi‐views modeling, the platform is a flexible framework and supports top‐down design, model reuse and interoperation, dynamic refinement of models, corporative design among different users in different stages, and the rebuilt of application rapidly.

This paper deals with high‐level models of the complex systems.

This platform helps to design, modeling, and simulation complex system (especially for weapon combat system). It can participate into all the stages of the development of complex product/system, and can support the validation, refinement, optimization of models, and systems.

This paper presents a multi‐views modeling approach for the modeling of complex system.

The engineering education accreditation (EEA) is a principal quality assurance mechanism. However, at many education institutions, the most labor-intensive work of EEA process is accomplished manually. Without the support of computer and information technology, the EEA process leads to high labor intensity, low work efficiency and poor management level. The purpose of this paper is to build a complex network model and realize an information management system of talent training program for supporting the EEA process.

Based on polychromatic graph (PG), this paper builds a network model of talent training program for engineering specialty. The related information and data are organized and processed in this network model. From the bidirections of top-down and bottom-up, the user requirements are retrieved automatically in logic layer. Together with the specialty of mechanical engineering, the proposed PG-based network modeling method is applied and the corresponding information management system is realized.

The study results show that the PG-based network modeling method takes full advantages of the strong simulation ability of PG to model the complex network system and has some unique merits in formal expression of problem, efficient processing of information and lightweight realization of system. Further, the information management system of talent training program can reduce the tedious human labor and improve the management level of EEA process dramatically.

This paper proposes a PG-based network modeling method, in which the nodes and the edges can be painted by some unified colors to describe the different kinds of activities and the various types of interactions. Theoretically, this modeling method does not distinguish the activities, the interactions and their properties in graphic symbol and the problem size is diminished about a half. Furthermore, this paper provides an effective experience and idea to the education institutions for implementing the engineering education accreditation, increasing the education management efficiency and promoting the talent training quality.

The purpose of this paper is to advance supply chain network theory by applying theoretical and empirical developments in complex network literature to the context of supply chains as complex adaptive systems. The authors synthesize these advancements to gain an understanding of the network properties underlying efficient supply chains. To develop a suitable theory of supply chain networks, the authors look to mirror the properties of complex network models with real‐world supply chains.

The authors review complex network literature drawn from multiple disciplines in top scientific journals. From this interdisciplinary review a series of propositions are developed around supply chain complexity and adaptive phenomena.

This paper proposes that the structure of efficient supply chains follows a “scale‐free” network. This proposal emerges from arguments that the key properties of efficient supply chains are a short characteristic path length, a high clustering coefficient and a power law connectivity distribution.

The authors' discussion centres on applying advances found in recent complex network literature. Hence, the need is noted to empirically validate the series of propositions developed in this paper in a supply chain context.

If efficient supply chains resemble a scale‐free network, then managers can derive a number of implications. For example, supply chain resilience is derived by the presence of hub firms. To reduce the vulnerability of supply chains to cascading failures, it is recognized that managers could build in redundancy, undertake a multi‐sourcing strategy or intermediation between hub firms.

This paper advances supply chain network theory. It offers a novel understanding of supply chains as complex adaptive systems and, in particular, that efficient and resilient supply chain systems resemble a scale‐free network. In addition, it provides a series of propositions that allow modelling and empirical research to proceed.

The purpose of this paper is to highlight how systems thinking contributes to decision making in uncertain contexts that are characteristic of service systems. Based on the assumption that service systems face complex conditions, the paper posits that systems thinking may support the understanding of key issues in service management.

This paper proposes an interpretation of complexity in the context of service systems, which highlights the perspective change that occurs when a systems approach is adopted. The offered conceptual perspective is then brought to an operational level, in spite of the complexity of the decisions driving a viable system, by modelling a service system as a network of agents, resources, processes and decisions through the use of fuzzy logic. The paper reviews service management research streams, and takes a deeper look at the concepts of service systems and complex service systems. The paper then proceeds to discuss how systems thinking contributes to service management by proposing a systems interpretation of complexity.

Service management theories and models may be enhanced by integrating prevailing approaches, based on a quantitative and mechanistic view of service systems dynamics, with systems thinking‐based meta‐models that can be used in better understanding service exchanges. The findings of the paper also show how the integration of an engineering approach can be insightful to the understanding of service systems; adopting a Viable Systems Approach (VSA) as a meta‐model can be useful in fully comprehending market behaviour in uncertain conditions.

The originality of this paper lies in exploring the contribution of systems thinking, in particular of the Viable Systems Approach (VSA), to service management and decision making.

The purpose of this paper is to propose a new system frame named ACPI for research and development (R&D) management of complex equipments according to the ideas of artificial societies, computational experiments, parallel execution and interactive optimization.

An artificial system which can effectively model, simulate and recur the main features and behaviors of a real R&D system of complex equipment is established at first. The structure and function of the system and its subsystems, and the relation of factors in the system are analyzed. Then one can perform computation experiment, modeling and simulation in the artificial system to obtain the optimal solutions. Finally, practice these solutions in a real system and at the same time perform the solutions in artificial system, forecasting and warning the possible new situations and problems in a realistic process, and provide controlling scheme.

The typical characteristics and solutions of the R&D system of complex equipment are analyzed. The sketch scheme of ACPI, the system frame of ACPI for R&D management of complex equipments are proposed, and the key technologies used in implementing ACPI of R&D system of complex equipment are studied.

The outcome of this paper can be used in computation experiments, management and optimization of R&D systems of complex equipment.

The sketch scheme of ACPI, the system frame of ACPI for R&D management of complex equipments are proposed first. The ACPI system can supply a high‐performance, open and interactive platform for the analog simulation and computation experiments of the R&D process management of complex equipment.

The purpose of this paper is to present a new concept based on a neural network validity approach in the area of multimodel for complex systems.

The multimodel approach was recently developed in order to solve the modeling problems and the control of complex systems. The strategy of this approach coincides with the usual approach of the engineer which consists in subdividing a complex problem to a set of simple, manageable sub‐problems that can be solved separately. However, this approach still faces some problems in design, especially in determining models and in finding the appropriate method of calculating validities.

A novel approach based on neural network validity shows very remarkable performances in multimodel for complex systems.

The validity of each model is based on the convergence of each neural network. For a fast convergence the proposed approach can be online to give a good performance in multimodel representation for system with rapid dynamics.

The proposed concept discussed in the paper has the potential to be applied to complex systems.

The suggested approach is implemented and reviewed with a complex dynamic and fast process compared to the residue approach commonly used in the calculation of validities. The results prove to be satisfactory and show a good accuracy.

Anticipating behaviour and responding to the needs of complexity and the problematic issues that they can generate requires modelling to facilitate analysis and diagnosis. Using arguments of anticipation as an imperative for inquiry, the purpose of this paper is to introduce generic modelling for living systems theory, and assigns the number of generic constructs to orders of simplex modelling. An nth simplex order rests in an nth order simplex cybernetic space. A general modelling theory of higher orders of simplexity is given, where each higher order responds to every generic construct involved, the properties of which determining the rules of the complex system being that is represented. Higher orders of simplexity also explain greater degrees of complexity relatively simply, and give rise to the development of new paradigms that are better able to explain perceived complex phenomena.

This is part 1 of three linked papers. Using principles that arise from Schwarz’s living systems set within a framework provided by cultural agency theory, and with a rationale provided by Rosen’s and Dubois’ concepts of anticipation, the papers develops a general modelling theory of simplex orders. It shows that with the development of new higher orders, paradigm shifts can occur that become responsible for new ways of seeing and resolving stubborn problematic issues. The paper is composed of two parts. Part 1 establishes the fundamentals for a theory of modelling associated with cybernetic orders. Using this, part 2 establishes the principles of cybernetic orders using simplex modelling. This will include a general theory of generic modelling. Part 3 extends this, developing a fourth order simplex model, and exploring the potential for higher orders using recursive techniques through cultural agency theory.

Cultural agency theory can be used to generate higher simplex through principles of recursion, and hence to create a potential for the generation of families of new paradigms. The idea of conceptual emergence is also tied to the rise of new paradigms.

The use of higher order simplex models to represent complex situations provides the ability to condense explanation concerning the development of particular system behaviours, and hence simplify the way in which the authors analyse, diagnose and anticipate behaviour in complex situations. Illustration is also given showing how the theory can explain the emergence of new paradigms.

Cultural agency can be used to structure problem issues that may otherwise be problematic, within both a top-down and bottom up approach. It may also be used to assist in establishing behavioural anticipation given an appropriate modelling approach. It may also be used to improve and compress explanation of complex situations.

A new theory of simplex orders arises from the new concept of generic modelling, illustrating cybernetic order. This permits the possibility of improved analysis and diagnosis of problematic situations belonging to complex situations through the use of higher order simplex models, and facilitates improvement in behavioural anticipation.

The purpose of this paper to present a practical and systematic approach to estimate the availability of a process plant using generalized stochastic Petri nets (GSPNs). The actual live problem at a fluid catalytic cracking unit (FCCU) of a refinery is used to demonstrate this approach.

A majority of models used for estimation of availability of a complex system are based on the assumptions that the failure of the system is associated with only a few states, and the system does not face different operating conditions, repair actions and common-cause failures. In reality, this is often not the case. Therefore, it is necessary to construct more sophisticated models without such assumptions. In this paper, an attempt has been made to model interaction of component failures, partial failures of components and common-cause failures.

The superiority of this approach over other modeling approaches such as fault tree and Markov analysis is demonstrated. The proposed GSPN is a promising tool that can be conveniently used to model and analyze any complex systems.

GSPN was used to model the reactor-regenerator section of FCCU, which is quite a large system, which shows the strength of modeling capability. The use of Petri nets (PNs) for modeling complex systems for the purpose of availability assessment is demonstrated in this paper. Sensitivity analysis was also carried out for various subsystem/components.

No similar work has been conducted for FCCU using GSPN as per literature incorporating different operating conditions and common-cause failures. The understanding and usage of PNs require a steep learning curve for the practitioners, and this paper provides an approach to estimate availability measures for the complex system.

The importance of context dependent modelling of complex systems, depending on the observer's point of view will be discussed. Thus, context is synonymous for the content of a problem in a frame of the goals, starting points and ways to achieve these aims. In this light, difficulties of model validation and a general method how to overcome them was discussed. The relations among subject – object – model in the light of a systems approach; Charles Sanders Peirce's triad principle and the semiotic principle of communication was presented.

The appropriateness of a system dynamics methodology, which is due to its transparency and clarity an excellent tool for modelling of complex systems.

In the paper the equivalence of different methodologies was shown, whose differences and similarities can be judged only in context of a problem and the aims of researches. For illustration, the methodology is applied to a tourism system, which possesses the typical properties of global and local organisations. A verbal description of a tourism problem is followed by a causal loop diagram, which helps to discuss the problem categorically.

As the methodology is implemented using quantitative model and POWERSIM tools; it offers the solution of national tourism strategy implication, selected from different scenarios.

This paper presents a simulation model of the tourism in a frame of system dynamics, developed from qualitative models, as an illustration of the discussed methodology.