Global Journal of Technology and Optimization

This paper is to investigate the current challenges, trends, relevance and the concept of supply chain performance measurements in developing nation basic metal industries. This paper gives an overview of different performance measurement tools that can be applied to measure and improve the supply chain systems of an organization. Since the study is focuses on supply chain performance measurements practices in the Ethiopian basic metal industries. The paper identifies and discusses the main motives and impact of supply chain performance measurement practice on the performance of basic metal manufacturing industries in Ethiopia and provided solutions to areas that needed improvement. While the study reviews the relevance of the main strategies, models and tools to measure the performance of supply chain in manufacturing industries is done though the research is conducted based on a qualitative methods. Structured questionnaires, field observation and oral interviews research techniques is use to get primary data from the sectors. Performance measure sets is identified through descriptive analysis. This study found that, large number of companies that now accomplish performance measurement systems at local and global level. Even though, most manufacturing firms use traditional performance measurement systems such as financial performance measurement systems. But this traditional performance measurement systems were not properly evaluates the performance and competitiveness of the whole supply chain systems of manufacturing sectors. Also according to this research proposes that the concept of supply chain performance measurement is not fully hold by the Ethiopian basic metal industries and highlights the difficulties associated with its implementation. Therefore, to tackle the challenges and attempt to the current performance measurement limitations, further studies need to involving other sectors and industries needs to be undertaken in order to gain an in-depth understanding of the key factors associated with the accomplishment of supply chain performance measurement practices in Ethiopia.

The Levenberg-Marquardt (LM) algorithm is the most commonly used training algorithm for moderate-sized feed forward artificial neural networks (ANNs) due to its high convergence rate and reasonably good accuracy. It conventionally employs a Jacobian-based approximation to the Hessian matrix, since exact evaluation of the Hessian matrix is generally considered computationally prohibitive. However, the storage of Jacobian matrix in computer memory is itself prone towards memory constraints, especially if the number of patterns in the training data exceeds a critical threshold. This paper presents a first attempt of evaluating the exact Hessian matrix using the direct differentiation approach for training a multilayer feed forward neural network using the LM algorithm. The weights employed for network training are initialized using a random number generator in MATLAB (R2010a). The efficiency of the proposed algorithm has been demonstrated using the well-known 2-spiral and the parity-N datasets, and the training performance has been compared with the Neural Network Toolbox in MATLAB (R2010a) which employs the conventional Jacobian-based learning methodology.