Comparative study of genetic and discrete firefly algorithm for combinatorial optimization

Abstract

Flexible job-shop scheduling problem (FJSP) is one of the most challenging combinatorial optimization problems. FJSP is an extension of the classical job shop scheduling problem where an operation can be processed by several different machines. The FJSP contains two sub-problems, namely machine assignment problem and operation sequencing problem. In this paper, we propose and compare a discrete firefly algorithm (FA) and a genetic algorithm (GA) for the multi-objective FJSP. Three minimization objectives are considered, the maximum completion time, workload of the critical machine and total workload of all machines. Five well-known instances of FJSP have been used to evaluate the performance of the proposed algorithms. Comparisons among our methods and state-of-the-art algorithms are also provided. The experimental results demonstrate that the FA and GA have achieved improvements in terms of efficiency. Solutions obtained by both algorithms are comparable to those obtained by algorithms with local search. In addition, based on our initial experiments, results show that the proposed discrete firefly algorithm is feasible, more effective and efficient than our proposed genetic algorithm for solving multi-objective FJSP.

An Imperialist Competitive Algorithm for a Real-World Flexible Job Shop Scheduling Problem

“…In order to compare the proposed ICA, implemented in (C++), we implemented (C++) the most applied method for the FJSP, a Genetic Algorithm (GA) (see Chaudhry and Khan [15], Amjad et al [16]). Moreover, a discrete Firefly Algorithm (DFA) showed good results for the FJSP (see Lunardi et al [5], Lunardi and Voos [4], Yang [17]). In this way, in order to check the performance of the continuous version, we implemented (C++) a continuous Firefly Algorithm (FA).…”

“…But when the number of jobs rises, it is difficult to find an optimal solution in a short time. Many researchers have proposed heuristic methods to solve the FJSP, such as genetic algorithm (GA) [4], firefly algorithm [5], artificial bee colony [6], particle swarm optimization [7], tabu search [8], and memetic algorithm [9].…”

A Mathematical Model and a Firefly Algorithm for an Extended Flexible Job Shop Problem with Availability Constraints

“…To solve the MILP models, we used the IBM ILOG CPLEX 12.7 solver with default parameters and a time limit of 3600 seconds. The DFA proposed in this work, and the Genetic Algorithm (GA) proposed in a previous work [7], were coded in C++. The MILP models, the DFA and the GA were run on an Intel Core i7 2.70 GHz, with 8 GB of RAM memory.…”

Saldırı Tespiti için Ateş Böceği Algoritması Tabanlı Özellik Seçim Yöntemi ve Yapay Bağışıklık Sistemi

“…Another experiment was achieved to solve the flexible job-shop scheduling problem (FJSP) with the Discrete Firefly Algorithm and multi-objective Genetic Algorithm (Lunardi & Voos, 2018). It was discussed that the proposed FFA was effective only for small instances of the problem.…”