Evolutionary algorithms and agricultural systems Mayer D. Kluwer Academic Publishers, Norwell, MA, 2001. 107 pp. Type: Book Reviews: 1 By: Ajith Abraham Optimization (G.1.6); Life And Medical Sciences (J.3); Sorting And Searching (F.2.2...); Biology And Genetics (J.3...) Performance, Theory, Algorithms Published: August 28, 2002 This enjoyable work is one of the smallest technical books I have read recently. The author used evolutionary computation for the optimization of an agricultural system project based in Australia. This work is a simply a compilation of the author’s previously published works. The book is well organized, with adequate references, and fulfils its purpose, even though it is short and simple. In chapter 1, Mayer introduces the reader to the complexity of a typical agricultural system using the case study of a beef property model for northern Australia. The various types of models used for study and optimization of agricultural systems are presented. A number of unsuccessful works, where conventional mathematical programming techniques (such as linear programming, dynamic programming, and mixed integer programming) have failed, and where global optimization techniques have been adopted, are pointed out. This is illustrated clearly in Table 1 of chapter 1. Detailed features and methodologies of agricultural systems models are given in chapter 2, emphasizing the need for multiobjective optimization algorithms. Chapter 3 serves as a tutorial, discussing the fundamental concepts of evolutionary algorithms, namely genetic algorithms and evolution strategies. This chapter also discusses the application of these algorithms to agricultural systems, such as greenhouse protection, farm management, and breeding programs. In chapters 4 and 5, some alternative optimization techniques, including gradient-based methods, direct search, simulated annealing, and tabu search, are introduced. The author discusses the relevance of these techniques to the optimization of agricultural systems. Some performance comparisons are made using test standard test functions. Empirical results also reveal the importance of the no free lunch theorem [1]. Chapter 6 addresses the importance of evolutionary algorithm parameter selection for optimal performance. Starting from the genotype representation of the problem domain, the importance of population size, selection of parents, replacement strategy, and reproduction operators (mutation and crossover) is clearly illustrated, with a focus on optimizing agricultural systems. Some future developments are also presented in chapter 7. As is evident from the technical content, most of the algorithms that the author has used in the optimization process are well-established techniques in the literature, widely adopted by academia and industry. In the concluding chapter, the author admits that systems researchers in the agricultural field tend to be late adopters of technology. Even though the users are late adopters, however, the book fails to reveal the latest developments in evolutionary computation and other search techniques. The author could have introduced some of the latest technical developments, such as co-evolution-based evolutionary algorithms and parallel evolutionary algorithms. The main focus of this book is on the importance of evolutionary computation for multiobjective optim ization problems, with a focus on agricultural systems. Even though the author includes success stories and different procedures for implementing evolutionary algorithms, the book does not provide any empirical comparison between evolutionary algorithms and other local search/global optimization techniques. A reader could appreciate the significance of evolutionary algorithms more if the author had provided comparisons between the different combinatorial optimization techniques as used on some agricultural systems, rather than illustrating the performance on test problems. The best feature of this book is its simple presentation style, which does not use much mathematics. I recommend this book as a tutorial for scientists and practitioners who would like to have a state-of-the-art research overview in the area of agricultural systems optimization. Review #: CR126413 (0211-0634) Review by: Ajith Abraham Copyright © 2000-2004 Reviews.com