Supplier selection and controlling using multivariate analysis

Purpose – An efficient supplier management is of central importance for successful supply chain management. It begins with the identification of potential suppliers and leads to the controlling of the supplier–buyer connection. This paper aims to design a new supplier-rating system that fits the practical needs of supplier rating and supports many other steps of the suppliers management process. Design/methodology/approach – An empirical study on supplier rating among 193 industrial companies revealed that the existing methods for supplier rating do not satisfy the needs in practice. This paper describes a multivariate analysis tool for managing a pool of engaged or future suppliers. A constructed ideal supplier serves as a reference to compare all suppliers by means of multivariate analysis methods. A case study illustrates the application of the tool. Findings – The empirical study found out new needs for supplier-rating methods such as graphical representation of the suppliers. The designed supplier-rating system uses principal component analysis to create a classification and ranking of the potential suppliers by means of ellipsoid clusters. If this new supplier-rating system is applied repeatedly, a dynamic observation of the suppliers is guaranteed, always corresponding to real market conditions. Originality/value – The new system has to run computerized and can be used for pre-qualification, selection and controlling of suppliers. It is easy to handle and practicable without using manual criteria-weighting.