Productivity plays a pivotal role in profitability and success of business. In this study, the wood cutting activity in Indian sawmills is selected. This study replicates the novel approach by integrating spherical fuzzy DEMATEL (Decision Making Trial and Evaluation Laboratory) and artificial neural networks (ANN) to improve the wood cutting productivity in Indian sawmills. The measure of betterness is selected as net productivity rate (NPR), a time-based labor productivity measure. The methodology unfolds in two crucial steps. First, SF-DEMATEL is employed to unearth influential factors affecting wood cutting, delving into their interrelationships through fuzzy logic. This process provides relationships between key determinants and their interconnected dynamics. Secondly, an ANN, a machine learning algorithm, is harnessed to predict wood cutting performance based on these identified factors. The ANN is trained using historical or simulation data, paving the way for predictions under diverse scenarios. The novelty of this approach lies in its holistic precision. The results showcase that lifting index and log weight emerge as primary influencers on productivity, with NPR, occupational risk index, and perceived exertion ranking lower. In the grand tapestry of factors, the study unveils universal driving forces, such as the weight of the log and lifting index. The ANN model, attaining a remarkable RMSE = 0.0478 and R2 = 0.9783 for training set and for training data and RMSE = 0.0487 and R2 = 0.9727 for testing data. This contributes to the comprehensive ranking comparison of factors derived from both Fuzzy DEMATEL and ANN. In summation, the fusion of Fuzzy DEMATEL and ANN unravels the intricacies of wood cutting dynamics. By identifying key factors and predicting performance, this approach provides a transformative gateway to enhance wood cutting quality and efficiency, thereby elevating the overall productivity of the woodworking industry.
Read full abstract