As manufacturing companies endeavor to meet market expectations and adjust to Industry 4.0, they face the challenge of producing high-quality products at a maximum production rate. Machine tools (MTs) utilized in various industries must maintain performance to guarantee precision and functionality throughout their run. However, the quality and productivity of MTs are significantly affected by uncontrolled heat flow within the MT components, mainly in the spindle. The transient heat flow within the complex geometry structures of spindle components, such as bearings and spindle motors, creates a complex thermal gradient that leads to nonlinear thermal deformation in the MTs. Inaccurate evaluation of heat sources can lead to the compromised performance of machine tool spindles. The present study introduces a comprehensive framework incorporating a 3D finite element model and an inverse method for estimating the three primary heat sources, including the spindle motor, using experimentally measured temperatures within the spindle housing with minimal sensors. This approach relies on directly measured temperature data to accurately assess thermal properties, bypassing the need for analytical calculations. The optimization process converges when the simulated temperature profiles closely align with the corresponding experimental data, yielding the respective heat flux values. Mean square error (MSE) values of 0.73, 2.7, and 0.36 are achieved for the front bearing, motor, and rear bearing, respectively. The experimentally measured temperature data from the spindle housing serves as the basis for this estimation process.