An adaptive firefly algorithm for blocking flow shop scheduling problem

摘要

In this paper, we present an Adaptive Firefly Algorithm (AFA) for solving the Blocking Flow Shop Scheduling Problem (BFSSP). It is known that the basic firefly algorithm (FA) works on continuous search space, while the BFSSP is a discrete problem. To handle discrete variables, the Smallest Position Value (SPV) rule is employed. An adaptive parameter strategy is utilised to reduce the dependence on parameters. Furthermore, two local search operators are used to improve the quality of solutions. To save computational time, a random attraction model is used to decrease the number of attractions among fireflies. Experiments are conducted on a set of Taillard's benchmark instances. Simulation results show that the proposed AFA achieves better solutions than the basic FA and four other algorithms.