TLBO Algorithm for Global Optimization: Theory, Variants and Applications with Possible Modification

Abstract

A large number of nature-inspired methods used in the area of Artificial Intelligence (AI), to
solve a wide range of problems. Computational Intelligence (CI) is one of the most specialized
areas of AI with the help of most nature-inspired techniques. Teaching Learning-based
Optimization (TLBO) algorithms are an important method of swarm intelligence (SI)
metaheuristics. Due to its noble features over other methods of SI, it was designed for a wide
variety of optimization problems. The algorithm was popular around many domains of global
optimization, such as electrical and electronics engineering, civil engineering, mechanical
design, artificial intelligence, physics, computer science, biotechnology and economics. The fact
that it only uses regular control parameters and does not need specific algorithms, is one of the
key features of an algorithm. It contributes trendy the direction of the algorithm’s effectiveness,
since it reduces the errors caused by excessive parameter tuning. The algorithm is based on the
natural process of learning the classroom. TLBO has recently gained a very high interest in
research from a variety of domains with large audiences. In this paper, therefore, several
researchers’ publications using TLBO were reviewed and summarized. Initially, an introductory
material on TLBO stays presented that explains the context of the natural structure and its
conceptual basis for optimization. In addition, the recent TLBO version is discussed in detail,
categorizing it in a new hybridized and multi-objective model. The main application of TLBO is
also systematically described. The paper provided an analysis of the TLBO algorithm and its
capabilities. This paper describes the comparison of TLBO variants also.