Application of an evolutionary algorithm-based ensemble model to job-shop scheduling

Choo Jun Tan, Siew Chin Neoh, Chee Peng Lim, Samer Hanoun, Wai Peng Wong, Chu Kong Loo, Li Zhang, Saeid Nahavandi
2017 Journal of Intelligent Manufacturing  
Saeid (2017) Application of an evolutionary algorithm-based ensemble model to job-shop scheduling. Journal of Intelligent Manufacturing. ISSN 0956-5515 (In Press) Published by: Springer URL: http://dx. Abstract In this paper, a novel evolutionary algorithm is applied to tackle job-shop scheduling tasks in manufacturing environments. Specifically, a modified micro genetic algorithm (MmGA) is used as the building block to formulate an ensemble model to undertake multi-objective optimisation
more » ... ms in job-shop scheduling. The MmGA ensemble is able to approximate the optimal solution under the Pareto optimality principle. To evaluate the effectiveness of the MmGA ensemble, a case study based on real requirements is conducted. The results positively indicate the effectiveness of the MmGA ensemble in undertaking job-shop scheduling problems. Keywords Multi-objective optimisation, evolutionary algorithm, ensemble models, job-shop scheduling 2 Heuristic-based Scheduling Problems Methods for solving scheduling problems can be broadly classified into two categories [38] . The first is enumeration-based methods that utilise dynamic programming or branch-and-bound techniques to find the optimal solution. The second is heuristic-based search methods that find near optimal solutions. The enumeration-based methods eliminate candidate solutions by employing restrictive criteria. However, it is well-known that the enumeration-based methods are time-consuming, and are not efficient for large-scale problems [62] , [49] . State space explosion as a result of increasing the number of state variables is the key limitation of dynamic programming techniques, owing to increasing number of state variables [62] . On the other hand, the long execution time owing to the number of variables involved in branching as well as the strategies used for bounding is the key limitation of branch-and-bound techniques [49] . In regards to heuristic-based search methods, the candidate solutions are produced by meta-heuristics searches, e.g. single-objective optimisation using constructive heuristics [42] . Another example is a hybrid genetic programming and hyper-heuristic method proposed in [41] . The method is used in a single-objective jobshop scheduling problem for evolving dispatch rules, whereby the results show the effectiveness of the genetic programming-based method over hybrid genetic algorithm (GA) methods.
doi:10.1007/s10845-016-1291-1 fatcat:etynw7ww5rhzrmd6nstryulyw4