Customer Reviews

Theory of Modeling and Simulation, Second Edition

5 star:		(3)
4 star:		(1)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

The first edition of this book appeared in 1976 and had only one main author, Bernard P. Zeigler, who authored other appreciated classic text books. The first edition was already a must read for the simulation community, which has been asking for an updating. The new edition is not just a re-edition with small corrections. The advances in simulation and modelling techniques helped to design a really new edition, necessary to meet the need for a theoretical foundation. Bernie Zeigler developed a unifying formalism DEVS (Discrete Event System Specification). With DEVS it is not only possible to specify discrete event systems; it is also particularly suited to integrate both continuous and discrete paradigms for modelling and simulation (discrete time & differential equations). DEVS features are also present in the systems theory.

The first theme of this book is to propose a sound methodology for the construction of complex system models based on the unifying concept of the DEVS Bus. This concept enables the mapping into DEVS of different models expressed in various formalisms such as Petri nets, Hierarchical block diagrams, cellular automata. The second theme focuses on parallel and distributed simulation of discrete event models. Distributed simulation has the potential to support the co-existence of multiple formalisms in multimodels.

I will try to depict the merits and demerits in comparison to other current and past books. First of all, being in the simulation field for ten years now, I have to cite some complementary references. Without giving an exhaustive list of all recent valuable books in the simulation field, we can find other high quality simulation text books ; for instance for continuous simulation [Cellier 1991], for general purpose simulation [Fishwick 1995] or even for more specific object-oriented simulations [Hill 1996]. However, I must admit that this Theory of Modelling and Simulation¡¨ is the only book available to help construct efficient object-oriented simulations of DEVS models on parallel and distributed environments. In addition, modelers who have to deal with HLA (High Level Architecture) compliant simulations will find in the book the reference to design integrative simulations.

The best significant features of this book are:

*The presentation of a comprehensive framework for continuous and discrete event modelling and simulation The exploration of the mathematical foundation of simulation modelling.

*The discussion of system morphisms for model abstraction and simplification *The presentation of a new approach to discrete event simulation of continuous processes. *The discussion of model credibility and interoperation * The highlighting of design concepts like such as modular and hierarchical model composition. *The presentation of a sound methodology to achieve parallel and distributed simulation models.

The reviewer only regrets the lack of public availability for DEVS platforms, despite the quotation of a web site, simulation environments (such as DEVSJAVA) are a real need to fully understand practical and theoretical aspects.

This book has multiple facets. From my teaching and scientific experience, I would say that the primary audience comprises engineers, scientists and other technical people working in academic institutions, but there are many chapters particularly well suited for students at a graduate or even undergraduate level. In my opinion, this was not the case of a previous book from the main author [Zeigler 1990] which was an excellent reference for specialists. I am now impressed by this ability to communicate at various levels. I strongly recommend this book for the reader interested in any kind of simulation and modelling, its impact on research in control will surely be recognized.

References: [Cellier 1991] CELLIER F.E., Continuous System Modeling, Springer Verlag, 1991. [Fishwick 1995] FISHWICK P.A., Simulation Model Design : Building digital worlds, Prentice-Hall, 1995. [Hill 1996] HILL D., Object-Oriented Analysis and Simulation, Addison-Wesley Longmann, 1996. [Zeigler 1990] ZEIGLER B.P., O-O Simulation with hierarchical modular models : Intelligent agents and endomorphic systems. Academic Press, London, 1990

This book have a bad title, must be "Theory of Modeling and Simulation with DEVS".

There are not a real Integration about Discrete Event and Continuous Complex Dynamic Systems. Only a Discret Event Systems Specification (DEVS), nothing more that DEVS. The book is anything but no visual simulation, no numerics management, no probabilities, no DAE-equations, no research operations, no object simulation, no ...

The DEVS concept born with the promise of join different formalism in the 70's (join discreet and continuous simulation), but yet the concept actually in the book is very primitive and not consider real numerical methods in only one layer (remember that many real commercial simulation software is not based in DEVS concepts. Why?, -Maybe, the DEVS concept have a rigorous and innecesary strictness in the framework concepts in modeling and simulation.)

For other hand, this book have bad thigs and good things.

Bad things:

- The algorithms presented are only fragments of pseudocode like C++ (or Java?). Where is all framework code?.
- The concepts are not in a today systemic/cibernetic vision. This is bad, because the definitions and concepts are yet of 70's. (the book is a second edition, where is the change?).
- Actually the DEVS is considered the corner stone of basic theory of discreet simulations according to autor and others, but in a book there are not real simulations for probe the theory.
- Where is THE continuous complex dynamic systems?.
- Lack of code production of real numerical and computacional methods.

The good things:

- If you don't know what is a DEVS this book is your resp.
- Is a source of excellent ideas, in special the study of Quantization and the Systems Design and Environments of M&S (IV Cap.).

Finally, the best of book is not the DEVS concepts (the book is 80% about DEVS), is the math-way how the Autor try define a basic Simulation Framework independent of the model.

Because the above, is a must have in a simulations books.

This book looks somewhat difficult but I think that it is the best book for whom researchs modelling and simulation methodology. Based on mathematics, most contents of it were explained rigorously. From the origin of its 1976 version, now in the 2'nd book, DEVS(discrete event system specification) formalism has been extended to cover the continuous state system as well. In addition to, the various extended versions of DEVS, such as Parallel DEVS, Real-time DEVS, Fuzzy DEVS, has been aloso dealt in it.

If you are familiar with the set-theory and the system theory (dealing with inputs and outputs) and looking for the simulation methodology from any IO system view, the book must be the best book to you.

My rating is not in relation to the contents of the book (I haven't been able to read it yet), it is just a warning for others to NOT buy this book in its Kindle edition.

Being a technical book, it has a lot of mathematical symbols, which are not displayed by Kindle!! only a little square is displayed instead of different symbols (less than, greater than, etc.). Also while using my computer to read it (Kindle for PC) NONE of the Figures are displayed!

With no figures and no symbols in a technical book, this kindle version is useless. I'm returning it and buying the paper version.

This book is an good example and reference how to solve optimization problems using simulation of the system (mostly it is Discrete Event simulation problems THERE ARE NO LINEAR OPTIMIZATION PROBLEMS IN REAL LIFE IN MY OPINION