In the batch processes, the transitional phase can be regarded as a connection process between its two neighboring stable phases. Thus, the effect on the transitional phase from its former phase to its latter phase exists. In this paper, a new transitional phase modeling and monitoring strategy is proposed. It provides two candidate models with respect to the effect of its two neighboring phases respectively. As a result, two candidate models are established for the transitional phase monitoring. To solve the above problem, an index is developed in this paper. The index is built with the combination of all the statistics and their corresponding control limits. In addition, as a key tool, it helps to select the suitable candidate model for every sample during the transitional phase. The reliability of the proposed strategy is illustrated with both the real three-tank experimental system and the Tennessee Eastman Process. (C) 2019 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.