Estimating the on-time probability for vendor selection problem

Customers expect fast delivery of products and services. Businesses understand this requirement and focus on efficient supply chains. The vendor selection process, which is complicated due a host of internal and external factors affecting the decision making, is fundamental to an efficient and responsive supply chain. As a selection criterion, the on-time probability for a vendor to supply a part can be used. In this paper, we have applied three quantitative methods, namely logistics regression, discrete time survival analysis and naïve Bayes classifier to evaluate a vendor. The mathematical models to estimate the on-time probability, were built and tested on a data set provided by a case company and evaluated with the help of key metrics.