Indonesia Conference Directory

1. Introduction In today-s competitive environment, vendor selection is considered as a strategic issue in the effective management of a supply chain. In most industries, the cost of raw materials and component parts represents the largest percentage of the total product cost. Therefore, selecting the right suppliers is the key to procurement process For many years, the traditional approach to supplier selection has been to select suppliers solely on the basis of price. Keeping in view the strategic importance of the supplier-s role in the functioning of supply chains the researchers have developed number of criteria, methods and models for supplier selection. This study involves a comprehensive review of different methods of supplier selection available in the literature and suggests a new approach based on statistical analysis on different characteristics of the supplier. 2. Methods The present study was carried out in Indian ferro-alloys industry to evaluate the suppliers- performance with respect to different quality characteristics of the raw material i.e. chrome ore. Data was collected in a planned manner on different quality characteristics of different lots supplied by different suppliers. An assessment of the existing quality status of Chrome ore, supplied by various suppliers, was done by multivariate analysis of variance (MANOVA). Duncans Multiple Range Test was used to classify the suppliers into homogeneous groups. Vendor rating was done on the basis of the obtained score using desirability function. Based on desirability value suppliers were ranked. 3. Results Results of multivariate analysis of variance shows there exist a significant difference in the quality characteristics of the different suppliers. Since different quality characteristics are having different specifications and are measured in different scales it is needed to convert them in a single 0-1 scale which is done using desirability function. For each supplier each quality characteristic was converted to a desirability value in the same scale. Combining the desirability values of each characteristics a single desirability value was obtained using geometric mean. This gives an index of the performance status of a supplier. Comparing these performance index the suppliers were ranked and the best one was selected. 4. Conclusions This study presents a structured methodology for supplier selection and evaluation based on statistical analysis considering all quality characteristics of all suppliers. This methodology can be applied for vendor selection and vendor rating for any industry irrespective of its nature of products and its characteristics. The use of desirability function converge all characteristics of the supplier into a single point in a 0-1 scale and desirability value of each supplier indicates its performance index in a 0-1 scale. Hence it has a vast scope of application. After selecting a suitable supplier its performance can be