A DEA-ANN framework based in Improved Grey Wolf Algorithm to evaluate the performance of container terminal.

Abstract

Managing the performance of port container terminals is one of the major challenges in the supply chain management. In response to this challenge, we propose a new framework to assess managers in evaluating the global performance of operations in port container terminals. The new framework integrates the Data Envelopment Analysis (DEA) and Artificial Neural Network (ANN). The DEA is used to compute the efficient score of the system. In ANN, we use the improved grey wolf optimizer based on Levy’s flights to improve learning. In order to prove the efficiency of our model, we apply the framework in 2 ports container terminal: Tangier Med Port and Casablanca Port. The result is compared to standard algorithms and outperforms the cited algorithm in order to avoid local minima. The new trainer improved grey wolf optimizer is also evaluated using four known classification datasets and on three approximation functions datasets.