Abstract
Determining the most suitable means of producing a given product and their ordering, or process planning, is traditionally done by manufacturing engineers who are experts in the internal processes and layout of a specific factory, and, with the exception of some limited support by Computer-Aided Process Planning tools, is largely a manual process. From the perspective of "manufacturing as a service", where the customer's product is not known in advance, the traditional approach has drawbacks: it requires human expertise to determine whether the product can be manufactured using the resources of a given service provider, and the small batch sizes (perhaps a single item) make the manual production of process plans uneconomic. To fully realise the manufacturing as a service vision, process planning must be automated, allowing service providers to 'bid' to manufacture products in real time. In earlier work, an approach was proposed to determine both whether a particular product is manufacturable given a set of available manufacturing resources, and how the product should be manufactured using those resources. In this paper, we present a tool that implements the definitions and algorithms in that work. We also link their abstract representations to concrete ISA-95 standards, and synthesise process plans suitable for execution by industrial manufacturing systems. The core of our tool is a new reasoner for orchestrating the activities of agents in the Evolvable Assembly Systems (EAS) architecture, an agent-based architecture for manufacturing control software designed to address rapidly changing product and process requirements.