Current industry trends show us that automating processes on the shop floor is becoming increasingly important; however, automation does not yet influence every production and production-related process. In particular, automating operations in the production planning department, ranging from production scheduling to documentation, is frequently overlooked. Work instructions for example are usually designed by hand which is a time-consuming process, especially in production situations with low-volume/high-variety products.

Trend towards more product customization
There is a trend toward more product customization, which implies the need for flexibility on the shop floor and challenges businesses to fulfil orders with short delivery times. Creating work instructions to meet this demand places a greater burden on office workers who design work instructions. The need to design work instructions in a short period of time is costly, and it can also lead to incorrect instruction content, which affects the process and product quality.

Student project
The development and implementation of a work instruction automating solution in a low-volume/high-variety assembly environment has proved to be important. For this reason, a graduation project at the HAN Lean QRM Centrum proposed a methodology for automatically generating work instructions for the shop floor from Computer-Aided Design (CAD) models. CAD files provide a wide range of product information. From this variety of information, geometric information was used to generate work instructions for assembly processes. This geometric information is extracted from the CAD file and processed to establish various assembly sequences. An optimization process was encountered to select an optimal sequence, which can be presented to the operator in form of work instruction. The proposed methodology was aiming to create work instructions for simple products, with elementary physical behaviour and simple assembly structure. The developed application was shown to be able to generate useful work instructions for the targeted product portfolio. The outcome of the application was tested in physical experiments using innovative Augmented Reality glass technology for the visualization of work instructions.

Experiments in the smart U-shaped assembly cell

The physical experiments were conducted at an experimental production facility, the smart U-shaped assembly cell built at the HAN University of applied sciences. During the experiment, operators were browsing through work instructions and assembled a variety of Lego products during the experiment. The digitalized work instruction format, provided by the proposed application is proven to be beneficial in assembly processes already, as it allows less-skilled personnel to carry out more difficult tasks while providing consistent quality during assembly. Furthermore, it allows production planning to present updated work instructions for the operators at the right place at the right time. These statements were supported by the conducted test case as well. To explore the possibilities for the standardization of the proposed solution the opportunities were studied to implement the system into the Industrial Internet Reference Architecture (IIRA).

Figure 1 – Work instruction visualization on an AR glass

Potential demonstrated but further research needed
The gathered knowledge during the project forms a strong base for further application development. The potential has been demonstrated, however, to fully utilize the benefits of the solution, further researches have to tune on this topic. Future development of the application is recommended to increase processing speed which is currently the bottleneck of the application. Further possible actions would be to extend the applicability of the solution to a broader range of products with more complex assembly structures and physical behaviour. Involving companies and encountering practical industry aspects from various departments, roles from operators to management would result in even greater applicability and user benefits.

More information
For more information on the experiments and research please contact

András Ligeti, researcher

András Ligeti

E andras.ligeti@han.nl