Object Oriented Simulation -- A modeling and programming perspective

The objective of this work is best described with the first sentence of the preface: “The primary goal of this book is to introduce students to the basic principles of object-oriented modeling, design and implementation of simulation models.” The book succeeds in it. It further states “For undergraduate students, another goal is to provide relevant material for easy understanding of object orientation and simulation”. As described further down in this review, the book does not completely fulfill this second objective.

When skimming the content of the book, I enjoyed that a programming language stance has been adopted. However a deeper read may leave one on his hunger. The author remains shy on telling the limitation of general purpose programming languages for simulation. After having read the book, it is not clear whether I will loose something in adopting a language different from OOSimL for teaching Object Oriented Simulation.

The overall objective of the book is quite ambitious, and this is reflected by its content somehow. Part 2 is titled “Object Oriented Programming with OOSimL”. It covers using OOSimL the notion of software, program structure, object, classes, thread, exception and many more in only 200 pages. Motivating each of these topics in an assertive manner cannot be other than difficult and challenging. This part of the book can be easily shrunk. For example, the chapter 4 titled “Programs and Software Development” is rather superfluous. It is so light that it cannot be useful. A discussion and an enumeration of the points where OOSimL is superior than general purpose languages would have definitely consolidated the whole story.

The third part of the book, titled “discrete event simulation”, is didactic and clearly written. It covers and illustrate the important topics of the field such as multi-servers, event list, priorities, resources, waiting, interruptions and specification of input. The car-washing motivation is used as a recurrent example. It also provide a nice and concise introduction to distribution and confidence interval. The notion of clock is missing however.

All that said, this book deserves to be considered for teaching the modeling and programming of discrete event based simulations. Even if the part related to object-orientation is not optimal, it may be easily completed with dedicated manuel.