A Tuned-NSGA-II to optimize the total cost and service level for a just-in-time distribution network

Title:

A Tuned-NSGA-II to optimize the total cost and service level for a just-in-time distribution network. Ashkan Memari, Abd. Rahman Abdul Rahim, Adnan Hassan, Robiah Ahmad. Neural Computing and Applications. 9 March 2016. Pp 1-15. DOI 10.1007/s00521-016-2249-0.

 

Abstract
Distribution network planning has attracted the attention of many studies during last decades. Just-in-time (JIT) distribution has a key role in efficient delivery of products within distribution networks. In modeling of JIT distribution networks, the most frequently applied objectives are related to cost and service level. However, evaluating the impact of simultaneously minimizing total costs and balance between distribution network entities in different echelons still rarely complies with the current literature. To remedy this shortcoming and model reality more accurately, this paper develops a multi-objective mixed-integer nonlinear optimization model for a JIT distribution in three-echelon distribution network. The aims are minimization of total logistics cost along with maximization of capacity utilization balance for distribution centers and manufacturing plants. A non-dominated sorting genetic algorithm-II (NSGA-II) with three different mutation operators namely swap, reversion and insertion is employed to provide a set of near-optimal Pareto solutions. Then, the provided solutions are verified with non-dominated ranked genetic algorithm (NRGA) as well. The Taguchi method in design of experiments tunes the parameters of both algorithms, and their performances are then compared in terms of some multi-objective performance measures. In addition, a genetic algorithm is used to assess Pareto optimal solutions of NSGA-II. Different problems with different sizes are considered to compare the performance of the suggested algorithms. The results show that the proposed solution approach performs efficiently. Finally, the conclusion and some directions for future research are proposed.

Keywords:

Optimization, Genetic algorithm, JIT, NSGA-II, Taguchi method

 

Link:

http://link.springer.com/article/10.1007%2Fs00521-016-2249-0

 

Back    Previous     Next