This paper demonstrates an approach for predicting and optimizing energy consumption in skid-steer mobile robots (SSMRs) conducting manufacturing tasks. This work is unique in that it considers the energy associated with real-time predictions of slipping in the SSMR and further considers a specific application in which the SSMR is operating in an inverted (climbing) configuration on metal surfaces with homogeneous properties. The approach is based on a dynamic model that provides estimates of SSMR slipping motion during simulation. The model is used to estimate the underlying components of energy and will serve as the tool for objective function evaluation. The approach will follow previous path optimization strategies, parameterizing the path to provide design parameters and using appropriate optimization tools. A method to select the desired trajectory prior to conducting a manufacturing task is demonstrated. This paper primarily focuses on a scenario in which a climbing SSMR maneuvers on a steel surface by means of magnetic-based tracks with strong adhering forces. For this case, the friction due to slipping represents the primary source of energy consumption. This implies that the path selection is the most important parameter for the optimization.