There are reports on the analysis of single type inclusions in steel by spark emission spectroscopy technology, but it is always a hard issue that complex inclusions analysis in steel by spark emission spectroscopy. The complex inclusions in steel are in two forms, one type inclusion containing another type inclusion or one type inclusion couples with another type inclusion. Nevertheless, distinguishing the isolated two types of inclusions and the complex inclusion is difficult by spark emission spectroscopy technology. In this paper, spark source original position distribution analysis (OPA) is used to analyze complex inclusions in the cross sections of a high railway wheel. The OPA technology can characterize the distribution of compositions and inclusions in a large area by the high-speed data acquisition and the analysis of element’s spectrum signal excited by continuous excitation on the scanning process. At the same exciting position where both Al₂O₃ inclusion spectrum signal and MnS inclusion spectrum signal exist, through sequential spectrum signal correlation between Al₂O₃ inclusion and MnS inclusion, the complex inclusions information can be obtained at the position corresponding to time point. According to a good linear relationship between the spectrum intensity over a threshold value and average area of inclusions, the area of Al₂O₃ inclusion and MnS inclusion in Al₂O₃/MnS complex inclusion are obtained. The sum of Al₂O₃ and MnS is the area of Al₂O₃/MnS complex inclusion at the same position. The scanning electron microscope (SEM) method is also used to analyze the Al₂O₃/ MnS complex inclusions in a limited area. The area of each complex inclusion by SEM corresponds to the normalized area of inclusion by OPA analysis, both the SEM and the OPA methods are in good consistency, and the linear correlation coefficient is better than 0.99. The results of the other two parts B₂ and B₃ of the cross-section of high-speed rail wheel measured by OPA for verification are also matched with the results obtained by SEM. In other words, in the analysis of small area Al₂O₃/MnS complex inclusions, the normalized results of OPA analysis can meet the characterization also. Moreover, because the OPA can analyze the whole area of a large sample, more large Al₂O₃/ MnS complex inclusions can be detected. The weak point is that the detection of leakage of large inclusions by SEM since the smaller testing area can be avoided. An effective method for the Al₂O₃/MnS complex inclusions analysis in large steel components by OPA is developed.