AutoPack 2.0 - Automatic packaging and assembly of electrical cables based on optimization and machine learning

Important results from the project

The goal of the project is to increase the degree of digitalisation and automation within product and production development and ensure that production and assembly aspects are taken into account in early phases. Within the project, we have developed a tool for optimal cable routing that automatically delivers routings that minimizes material consumption. The tool has been connected in a framework for ergonomic simulation so that ergonomic aspects during assembly can be balanced against other functional requirements (such as material consumption and cost).

Expected long term effects

The project´s objective was to save 50% of the engineering hours spent on cable installations (which today is more than 10,000 hours/year) and to reduce the lead time by 25% for the installation work. The project has shown a 60% reduction in engineering hours and a 15% reduction in lead time. Today the developed framework does not take into account all aspects of cabling, which is why some manual work is still required after the automated process. Continued development of the framework will increase the quality of the installations and thus reduce the lead time further.

Approach and implementation

The project has been led by LiU and the other partners were Fraunhofer-Chalmers Center and Volvo Car Corporation. The project was divided into 6 work packages (APs) where AP1 set the requirements, AP2-5 developed different methods and tools which were validated in AP6. Overall, the project has worked well with structured follow-up, good communication between the project partners and the project has found a balance between the different partners´ goals. Some redistribution between the APs has been done, where time from AP5 was redistributed to AP2 and AP3.