Energy demand in production systems: A Queuing Theory perspective

Production systems are usually organised in departments consisting of machines, each one characterised by specific patterns of energy demand over time. This work proposes an analytical approach, based on the application of Queuing Theory, to model the power request and the consequent energy use in a production system. Despite the industrial context addressed, the model may be easily applied to small units (e.g., civil buildings) and other energy sources (e.g., thermal energy), thus giving more relevance to the approach proposed. The model can efficiently support green-field cases, particularly avoiding or integrating the traditional assumptions, such as load and coincidence factors (usually employed to determine the contractual electrical power), which provide a static view of the power needs of the system. In fact, the proposed queuing model considers the arrivals as the statistical distribution of the switch-on of machines and service completions as the statistical distribution of the processing times at the machines themselves, thus offering a dynamic view of the power loads. Therefore, the model may be helpful while assessing the contract with the energy supplier or planning the production schedule of plants with significant energy-related constraints, including plant services. A numerical example shows the application of the proposed approach and its results are compared to those determined by the traditional design methodology.