Powertrain Assembly Lines Automatic Configuration Using a Knowledge Based Engineering Approach

Technical knowledge and experience are intangible assets crucial for competitiveness. Knowledge is particularly important when it comes to complex design activities such as the configuration of manufacturing systems. The preliminary design of manufacturing systems relies significantly on experience of designers and engineers, lessons learned and complex sets of rules and is subject to a huge variability of inputs and outputs and involves decisions which must satisfy many competing requirements. This complicated design process is associated with high costs, long lead times and high probability of risks and reworks. It is estimated that around 20% of the designer’s time is dedicated to searching and analyzing past available knowledge, while 40% of the information required for design is identified through personally stored information. At a company level, the design of a new production line does not start from scratch. Based on the basic requirements of the customers, engineers use their own knowledge and try to recall past layout ideas searching for production line designs stored locally in their CAD systems [1]. A lot of knowledge is already stored, and has been used for a long time and evolved over time. There is a need to retrieve this knowledge and integrate it into a common and reachable framework. Knowledge Based Engineering (KBE) and knowledge representation techniques are considered to be a successful way to tackle this design problem at an industrial level. KBE is, in fact, a research field that studies methodologies and technologies for capturing and re-using product and process engineering knowledge to achieve automation of repetitive design tasks [2]. This study presents a methodology to support the configuration of powertrain assembly lines, reducing design times by introducing a best practice for production systems provider companies. The methodology is developed in a real industrial environment, within Comau S.p.A., introducing the role of a knowledge engineer. The approach includes extraction of existing technical knowledge and implementation in a knowledge-based software framework. The macro system design requirements (e.g. cycle time, production mix, etc.) are taken as input. A user driven procedure guides the designer in the definition of the macro layout-related decisions and in the selection of the equipment to be allocated within the project. The framework is then integrated with other software tools allowing the first phase design of the line including a technical description and a 2D and 3D CAD line layout. The KBE application is developed and tested on a specific powertrain assembly case study. Finally, a first validation among design engineers is presented, comparing traditional and new approach and estimating a cost-benefit analysis useful for future possible KBE implementations.