Accurate knowledge of temperature distribution in mechanical systems such as gearboxes is often addressed with extensive multiphysics simulations, because direct measurement during operation is challenging, and the full thermal system, including lubrication and surroundings, must be represented. This contribution presents Theta, a software tool based on the thermal network method (a lumped-parameter method), for efficient heat-transfer analysis in mechanical systems. Although thermal networks are well established, their application to complex mechanical systems has been limited by the manual effort required to build detailed and reliable thermal network models. Theta automates these time-consuming steps. Via a graphical user interface, users provide simple geometric inputs and model parameters (operating conditions, materials, and heat sources) to obtain temperature distributions and heat fluxes for the complete system, either transiently or in steady state. Evaluating internal heat fluxes supports a deeper understanding of the system’s thermal paths, enabling design modifications to improve thermal performance. Additionally, accurate knowledge of component temperatures supports material selection and can serve as inputs for improved downstream simulations, such as the prediction of friction and wear. Theta also features an optimization-based calibration workflow that uses user-supplied temperature measurements to automatically refine the thermal network model. Using worm gearboxes as an example of a multi-component mechanical system with coupled conduction and convection paths, we show that Theta generates accurate temperature distributions with low modelling effort and computation times in the order of 15 s per operating point. We validate Theta’s temperature predictions against measurements of more than 20 housing and component temperatures, measured over several hours across multiple operating conditions on different gearboxes. The results show that Theta can reliably predict temperature distributions and their evolution in gearboxes, enabling informed material and component selection as well as rapid parameter studies.