In the manufacturing sector, despite the vital role it plays, the consumption of energy is rarely considered as a manufacturing process variable during the scheduling of production jobs. Due to both physical and contractual limits, the local power infrastructure can only deliver a finite amount of electrical energy at any one time. As a consequence of not considering the energy usage during the scheduling process, this limited capacity can be inefficiently utilised or exceeded, potentially resulting in damage to the infrastructure. To address this, this thesis presents a novel schedule optimisation system. Here, a Genetic Algorithm is used to optimise the start times of manufacturing jobs such that the variance in production line energy consumption is minimised, while ensuring that typical hard and soft schedule constraints are maintained. Prediction accuracy is assured through the use of a novel library-based system which is able to provide historical energy data at a high temporal granularity, while accounting for the influence of machine conditions on the energy consumption. In cases where there is insufficient historical data for a particular manufacturing job, the library-based system is able to analyse the available energy data and utilise machine learning to generate temporary synthetic profiles compensated for probable machine conditions. The performance of the entire proposed system is optimised through significant experimentation and analysis, which allows for an optimised schedule to be produced within an acceptable amount of time. Testing in a lab-based production line demonstrates that the optimised schedule is able to significantly reduce the energy consumption variance produced by a production schedule, while providing a highly accurate prediction as to the energy consumption during the schedules execution. The proposed system is also demonstrated to be easily expandable, allowing it to consider local renewable energy generation and energy storage, along with objectives such as the minimisation of peak energy consumption, and energy drawn from the National Grid.