Planning Machine Activity Between Manufacturing Operations: Maintaining Accuracy While Reducing Energy Consumption

There has recently been an increased emphasis on reducing energy consumption in manufacturing. This is largely because of fluctuations in energy costs causing uncertainty. The increased competition between manufacturers means that even a slight change in energy consumption can have implications on their profit margin or competitiveness of quote. Furthermore, there is a drive from policy-makers to audit the environmental impact of manufactured goods from cradle-to-grave. The understanding, and potential reduction of machine tool energy consumption has therefore received significant interest as they require large amounts of energy to perform either subtractive or additive manufacturing tasks.

One area that has received relatively little interest, yet could harness great potential, is reducing energy consumption by optimally planning machine activities while the machine is not in operation. The intuitive option is to turn off all non-essential energy-consuming processes. However, manufacturing processes such as milling often release large amounts of heat into the machine's structure causing deformation, which results in deviation of the machine tool's actual cutting position from that which was commanded, a phenomenon known as thermal deformation. A rapid change in temperature can increase the deformation, which can deteriorate the machine's manufacturing capability, potentially producing scrap parts with the associated commercial and environmental repercussions. It is therefore necessary to consider the relationship between energy consumption, thermal deformation, machining accuracy and time, when planning the machine's activity when idle, or about to resume machining.

In this paper, we investigate the exploitability of automated planning techniques for planning machine activities between subtractive manufacturing operations, while being sufficiently broad to be extended to additive processes. The aim is to reduce energy consumption but maintain machine accuracy. Specifically, a novel domain model is presented where the machine's energy consumption, thermal stability, and their relationship to the overall machine's accuracy is encoded. Experimental analysis then demonstrates the effectiveness of the proposed approach using a case study which considers real-world data