Next procurement prediction using Markov Chain Model with Maximum Likelihood, Maximum Likelihood with Laplace Smoothing, Bootstrap and Maximum a Posteriori methods: Case of Kimia Farma, Ltd.

The aim of this study is to explore whether captured procurement data of Kimia Farma, Ltd. can be used to predict the next procurement using Markov Chain method accurately. Similar techniques have been implemented as reported by several articles. However, the accuracy varies. Therefore, this study explored the prediction using Markov Chain model with Maximum Likelihood, Maximum Likelihood with Laplace Smoothing, Bootstrap Approach, and Maximum a Posteriori. The intention of using multi-Markov’s approaches is to find the one that gives the best prediction. The results show that the prediction gave a little accuracy to predict next procurement activity, even though in one of the conditions, the prediction showed quite good results. Laplace Smoothing was considered as the best method for improving prediction results.