DOIONLINE 
DOIONLINE Publish In |
International Journal of Management and Applied Science (IJMAS)-IJMAS |
 Journal Home Volume Issue |
||||||||
Issue |
Volume-7,Issue-12 ( Dec, 2021 ) | |||||||||
Paper Title |
Statistical Model to Predict the Performance of Manufacturing Modules in Sri Lankan Apparel Industry | |||||||||
Author Name |
Nawodh W.A.S., Lokupitiya R., Samayawardena D. | |||||||||
Affilition |
Department of Statistics, Faculty of Applied Sciences, University of Sri Jayewardenepura, Gangodawila, Nugegoda, Sri Lanka ABC (Pvt) Ltd, Sri Lanka | |||||||||
Pages |
5-8 | |||||||||
Abstract |
With the development of technology and increasing competition between manufacturing firms in the global market, it has become essential for organizations to transform the manufacturing process to predictive manufacturing. Especially in Sri Lankan apparel industry, with increasing complexity in manufacturing modules as well as to keep the competitiveness against other global players, going towards predictive manufacturing is crucial. One of the key elements of predictive manufacturing is to identify the lower performing manufacturing modules in the factory so that management can take proactive actions to improve the performance of this volatile environment. As a solution, this study was conducted to develop a statistical predictive model to predict whether a manufacturing module will achieve its given target at the end of the day. Since the depended variable is binary, LogisticRegression model, Naïve Bayesianand Support Vector Machine algorithms were chosen. Based on domain knowledge expertise,predictor variables for the study were identified. To develop the model, data form five months were collected from the chosen factory and analysis was carried out using the R programming language. Collected data of the factory consisted with class imbalance problem and due to this, Mathews Correlation Coefficient was used for model evaluation. Through model improvement procedures, we were able to improve the model performance and final overall accuracy of the fitted model was 78%. Keywords - Performance prediction, Logistic Regression, Manufacturing Prediction, Data Mining | |||||||||
| View Paper | ||||||||||
