Novel Approach to Strength Modeling of Concrete under Triaxial Compression

Gene Expression Programming Approach to Cost Estimation Formulation for Utility Projects

“…Although GP applies most of key ideas of GA and uses GA operators such as selection, crossover and mutation with slight modifications, its nonlinear structure creates a more versatile system of representation than that of GA (Ferreira 2006;Gandomi et al 2012). GP produces computer programs with dynamic variability and hierarchical character presented in form of parse trees (Koza 1992).…”

“…A population member in hierarchically structured tree-based GP composes of functions and terminals selected from a set of functions and a set of terminals. Figure 1 is an illustration of a simple tree-based GP model (Gandomi et al 2012). GP can be implemented using any programming language (like LISP) capable of working with computer programs as data and linking, compiling and executing new programs (Koza 1992).…”

“…The GEP consists of five main components: 1) function set; 2) terminal set; 3) fitness function; 4) control parameters; and 5) termination condition (Gandomi et al 2012).…”

“…An advantage of the GEP technique is that the creation of genetic diversity is extremely simplified because genetic operators work at the chromosome level (Gandomi et al 2012).…”

“…GP's ability to develop simple prediction equations with no need to considering an existing relationship is its main superiority over the conventional statistical and ANN techniques (Gandomi et al 2012). GP is intrinsically capable of finding the best solution by evaluating fitness of the computer programs over numerous generations; on the contrary, as mentioned earlier, in ANN, data should be normalized at the outset, and best network architecture first should be established (Gandomi, Roke 2015).…”