Installing large floating objects during offshore operations is a challenging and failure-prone task, especially when passing through the splash zone due to extreme lifting loads on the wire and the payload. For a safe operation, it is essential to predict the peak loads on the installation system and create an early decision-making scenario for the installation vessel before starting the real operation on site. To this end, the extreme loads that can lead to unsatisfactory performance of the system must be evaluated accurately; however, the operation involves a great deal of uncertainty and physics complexity that can lead to unreliable decision-making. It is also challenging to perform numerical calculations to support ongoing marine operations, as it usually takes hours to evaluate different environmental load cases. Thus, it is essential to create an efficient prediction method associated with the environment and the corresponding response levels. In this study, a model is proposed that integrates physics-based simulations with Gaussian Processes, for estimating peak loads in lifting wires. The model offers the advantage of addressing shorter simulation times while still maintaining accuracy in predicting extreme response levels and quantifying the loads uncertainty during the operation. Bayesian Inference is used to incorporate the uncertainty, estimating hyper-parameters and predict the peak loads for various marine environmental conditions. A real case study is considered to demonstrate the application of the proposed model. The results show good agreement with the simulations obtained from time-domain dynamic analysis. The current study provide insights for both onboard and pre-planned decision-making on installation conditions, thereby enhancing predictive accuracy and improving safety in complex marine lifting operations.