Container reshuffle is one of the main problems that container terminals face for several reasons. One reason for container reshuffle is uncertain transaction type. Yard planner needs the information for the transaction type to allocate inbound containers without causing a reshuffle. The vessel agent submits the transaction type information on the discharge list. However, before the vessel's arrival, circumstances -such as change of the cargo owner or lack of information -are encountered; therefore, information on the discharge list is unreliable. Yard planner can know the exact transaction type only before the container exits. This article follows the given steps of the cross industry standard process for data mining (CRISP-DM) at a seaport in Turkey to predict the transaction type before vessel arrival. We propose a multiple logistics regression model integrated with the terminal operating system to provide sustainable outputs to planners. The model predicts the container transaction type with 89% accuracy.