Abstract: For the batch transportation of various types and grades of oils through product oil pipelines, accurately positioning batch interfaces is crucial to improving the operational quality of the pipelines. This paper presents a new method to achieve this purpose. An online monitoring system was implemented to collect near-infrared (NIR) spectra at the interfaces between No. 0 diesel and No. -35 diesel transported through a pipeline. Subsequently, a parametric logistic equation was used to analyze the variation trends in these spectra. The samples of NO. 0 dieselr, NO. -35 diesel, and their mixtures were then employed to establish a partial least squares (PLS) quantitative model of the NIR spectra for the cold filter plugging points (CFPP) of No. 0 diesel and No. -35 diesel mixtures. This model was utilized to predict CFPPs at the interfaces where these two grades of oil mix. The results demonstrate that the variation trends in NIR spectra during batch transportation align well with the parametric logistic equation. Samples of No. 0 diesel and No. -35 diesel, as well as their mixtures, exhibited wide ranges of CFPPs, and the NIR spectra showed a clear nonlinear relationship with the CFPPs. Partial least squares (PLS) NIR quantitative modeling in segments can effectively predict the CFPPs during batch transportation. The deviations between the NIR-based CFPPs and the measurements based on the laboratory standard method did not exceed 3 ℃, remaining within the reproducibility error limit required for the CFPP standard method in SH/T 0248. The proposed NIR spectroscopy online analysis approach is effective for the online monitoring of CFPPs as a quality indicator for the interfaces between No. 0 diesel and No. -35 diesel during the transportation process. This supports its application in assessing the interfaces of different grades of the same oil products transported through product oil pipelines. This method provides reference and supporting data for oil separation during the batch transportation process, while also reducing the workload for manual inspection.