The performance of product-driven manufacturing control: An emulation-based benchmarking study

Abstract

International audienceProduct-driven control may enable manufacturing companies to meet business demands more quickly and effectively. But a key point in making this concept acceptable by industry is to provide benchmarking environments in order to compare and analyze their efficiency on emulated large-scale industry-led case studies with regard to current technologies and approaches. In this paper, a benchmarking protocol is defined, in order to provide R\&D practitioners with benchmarking services in a product-driven implementation project. A component-based generic architecture is proposed to support this protocol, enabling to model and compare various control architectures. This benchmarking protocol is applied to an automotive-industry case study in order to evaluate the impact of making the products interact with the local decision centers. Finally the experiments show that product-driven control can perform as good as traditional centralized control, and that its robustness depends mainly of the local decision-making processes

Benchmarking flexible job-shop scheduling and control systems

“…The use of the platform Emulica (Pannequin et al, 2009), presented in the survey part of this paper, is a very interesting open simulation environment. This should facilitate future comparisons of control strategies since it would not force researchers to develop a new simulator each time a new contribution is under consideration, leading them to deploy their effort rather on control strategy instead of the simulation of the operating system.…”

“…Cavalieri, Macchi, and Valckenaers (2003) proposed a web simulation testbed for the manufacturing control community. Pannequin, Morel, and Thomas (2009) proposed an emulation-based benchmark case study devoted to a productdriven system, and Mönch (2007) proposed a simulation benchmarking system. These benchmarks are interesting since they try to deal with the dynamic behavior of the system to be controlled, which is harder to formalize in a simple and exclusively quantitative way, like benchmarks from the OR community.…”

An Evolutionary Simulation-Optimization Approach to Product-Driven Manufacturing Control

“…Hence, intelligent product driven manufacturing control emerges as a promising declination of multi-agent HMSs, and is actually defined by Pannequin [13] as a specialization of holonic agent-based distributed control paradigm where agent technology brings forward new fundamental insights on decentralized coordination and auto-organization, enabling new manufacturing decision-making policies and on-thefly reconfiguration capabilities and infotronics technologies address the issue of synchronization between physical objects and their informational representation. Beside, Trentesaux and Thomas [14] arguing that product-driven control is based on the assumption that the product is the core object in the design, manufacturing, logistic and services systems, define this paradigm as "a way to optimize the whole product life cycle by dealing with products whose informational content is permanently bound to their virtual or material content and, thus, are able to influence decisions made about them and participating actively to different control processes in which they are involved throughout their life cycle".…”

Intelligent distributed production control

“…In order to insure consistency of decision-making, more pragmatic approaches based on hybrid control combining the predictability of the centralized control with the agility and robustness against disturbances of the heterarchical control have been designed, allowing the integration of optimization models with features of cooperation and local autonomy. To name a few, the concepts of Holonic Manufacturing Systems (HMS) (Babiceanu and Chen 2006), Product Driven Systems (PDS) (Pannequin et al 2009) and Agent-Based Manufacturing (Maturana et al 1999) have been proposed to design future manufacturing systems. These concepts argue that the products-and more globally, all the production resources-can be modeled as an association between two parts: a physical part and an informational one.…”