Methodische Vorgehensweise für die Austaktung einer Modell-Mix Montagelinie in einem Baumaschinenunternehmen

Abstract

A key challenge in the modern manufacturing industry is the increasing complexity of production planning due to a high number of variants and an increasing number of changes in demand. This leads to process and time fluctuations and, in turn, to uneven employee utilisation. Production processes - in this work the assembly process in the construction machinery industry - must therefore be designed flexibly in order to adapt to this. The difficulty lies in achieving the highest possible worker utilisation in a multi-variant assembly -also known as model-mix assembly-, despite fluctuations in demand, while at the same time ensuring the planned line productivity (planned daily quantity). For this reason, line optimisation and with it the topic of cycle time coordination is becoming increasingly important. Due to the topicality of this subject and as a supplement to the existing literature, this scientific paper deals with the cycle time synchronisation of a model-mix assembly line. Approaches and methods for the actual line balancing as well as the underlying line balancing process are analysed. The aim is to find out how a fast and low-resource (i.e. time and employee requirements) line balancing process can be ensured, and how a high worker utilisation rate can be achieved while maintaining 100% line productivity. For this purpose, in addition to existing literature, real company data from the construction machinery manufacturer Wacker Neuson is used and analysed, and potential for improvement is subsequently identified and reviewed. The results have shown that a clear structure, but, above all, the introduction of an assembly logic, increases the efficiency of the line balancing process. This means that fewer resources (employees) and process steps are required and it is possible to react more quickly to changes in model mix or quantities. The aim would be to introduce an assembly priority graph for all models and variants. This would allow for faster, more transparent and sustainable scheduling without priority errors. Various approaches were identified to ensure high worker utilisation while maintaining the planned line productivity. One of the most important findings was that the usage probability should not be taken into account when calculating the number of variants and options, as this could obscure work content and lead to unplanned system fluctuations. Alternatively, quantiles should be used to analyse target times instead.

Date of Award	2024
Original language	German (Austria)
Supervisor	Herbert Jodlbauer (Supervisor)