Industrial companies are usually characterized by a immense demand of electrical and thermal energy as by a high requirement for reliability. The industrial microgrid designates a local supply concept that takes the respective framework conditions of industrial companies into account and involves local energy sources as well as energy storage units. Its goals are, on the one hand, to reduce supply costs and, on the other hand, to increase the supply reliability. In extension of the state of the art, this thesis focuses on the electrical energy storage operation in an industrial microgrid, where the combination of the individual cost-saving options stands in the foreground. On the basis of a model predictive control, an operating strategy is presented which limits the demand peak power, increases the self-consumption, maximizes the storage reserve for power outage ride through, and supplies primary control reserve. Since this requires a prediciton, established prediction methods are compared, and application-specific features and enhancements are presented. The result of this thesis shows that the presented strategy has the potential to increase the cost savings significantly. However, the achievement of the goals requires a complex energy management system and depends on various influencing factors. A detailed description of those factors as well as options which reduce the risk of non-optimal results are presented. Although supply tasks for industrial companies are the focus of this work, the results can be applied to those with similar framework conditions as well, like large building complexes, universities or hospitals for instance.