Purpose - Current major roadmapping efforts have all clearly underlined that true industrial sustainability will require far higher levels of systems' autonomy and adaptability. In accordance with these recommendations, the Evolvable Assembly Systems (EAS) has aimed at developing such technological solutions and support mechanisms. Since its inception in 2002 as a next generation of production systems, the concept is being further developed and tested to emerge as a production system paradigm. The essence of evolvability resides not only in the ability of system components to adapt to the changing conditions of operation, but also to assist in the evolution of these components in time. Characteristically, Evolvable systems have distributed control, and are composed of intelligent modules with embedded control. To assist the development and life cycle, a methodological framework is being developed. After validating the process-oriented approach (EC FP6 EUPASS project), EAS now tackles its current major challenge (FP7 IDEAS project) in proving that factory responsiveness can be improved using lighter multi-agent technology running on EAS modules (modules with embedded control). The purpose of this paper is to detail the particular developments within the IDEAS project, which include the first self re-configuring system demonstration and a new mechatronic architecture. Design/methodology/approach - The paper covers the development of a plug & produce system for FESTO AG. The work covers the background methodology and details its constituents: control system, architecture, design methodology, and modularity. Specific detail is reserved for the configuration approach which integrates several tools, and the commercially available control boards. The latter have been specifically developed for distributed control applications. Findings - The paper details probably the first self-configuring assembly system at shop-floor level. This is one of the very first industrial plug & produce systems, in which equipment may be added/removed with no programming effort at all. Research limitations/implications - The paper reports the findings and development carried out within the framework of a single project. It also clarifies that the solution is not a general panacea for all the issues within assembly. Practical implications - The implications are quite large as the work proves the validity of an approach that could change our way of designing and building assembly systems. In the words of an industrial partner, this is "a new way of engineering assembly systems". Social implications - Should this approach be used in industry then the implications could be huge. It would, for example, mean that new services are created, whereby assembly system modules are leased to users through a network of depots, rather than bought at a high cost. The system modules also have a far longer lifespan, implying very good ecological solutions. Originality/value - The highly original paper describes what is probably one of the very first projects to show that distributed control at shop-floor level is viable and technologically feasible.
