Intelligent Prediction Model: Optimized Neural Network for Lean Manufacturing Technology

  • Jobin M V Professor, Mets School of Engineering Mala
  • Aiswarya Menon Assistant Professor, Mets School of Engineering Mala


Lean manufacturing (LM) is a method, which focus on reducing wastes and increasing the productivity within manufacturing firms. Several analyses were performed on LM technology depending on minimal lead times, enhanced quality and reduced operating costs. However, limitation exists in understanding its role to develop managing commitment, worker involvement and in turn its organizational performance. This paper intends to propose a new Neural Network (NN) based intelligent prediction framework. The initial process is manual labeling or response validation, which is carried out by utilizing the responses attained for each questions under each factors including lean awareness, employee involvement, management commitment, lean technology, Organizational Performance (OP) and Organizational Support (OS). Subsequently, NN is exploited for prediction process, where the features (received responses) are given as input and the labeling values attained are set as target. Further, in order to improve the prediction performance, the NN training is performed by a new Hybrid Particle Swarm and Pigeon Optimization (HPS-PO) algorithm via tuning the optimal weights. In fact, the proposed algorithm is the combination of Particle Swarm Optimization (PSO) and Pigeon Optimization Algorithm (POA), respectively. Finally, the performance of the proposed model is examined over conventional methods in terms of prediction analysis and error analysis.