35 th International Conference on Computers and Industrial Engineering 543 DESIGN OF TEAM ORIENTED MIXED MODEL ASSEMBLY SYSTEMS M. Bulent Durmusoglu 1 , Murat E. Unal 1 and Suat Didari 2 1 Industrial Engineering Dept., Istanbul Technical University, Macka, Istanbul, Turkey 2 Uzel Makina Sanayi, A. S., Topcular, Rami, Istanbul, Turkey Abstract: Team oriented approaches are widely being used in modern assembly systems. However contributions to this research area in literature have not reached to the level it deserves. In this paper, the developed design methodology of assembly systems considering both team work and mixed model products is explained. The methodology consists of five steps: horizontal balancing, vertical balancing, creating physical station with the aid of scheduling under limited resources, product sequencing, and finally worker transfer system with the help of assembly line simulation. This methodology was started to be implemented step by step in a real life mixed model tractor assembly system. Key words: Mixed model balancing, teams, scheduling under limited resources, product sequencing, simulation 1. Introduction The classical assembly balancing for both single and mixed models is mostly based on horizontal balancing that many scientific researches have appeared in the literature (Bukchin et al., 2002, Erel & Gokcen 1999; Khan & Day 2002, Sparling & Miltenburg 1998). Merengo et al. (1999) separate the balancing problem into horizontal and vertical balancing. In these classical assembly balancing approaches, only one worker is assigned to an assembly station. However, team oriented approaches are widely being used in modern assembly systems. There are many advantages for team based assembly systems. Bukchin et al. (1997) describes them systematically. The main advantage of the approach is to create a team based worker environment for continuous improvement. The flow time is shortened due to the overlapping assembly tasks. The other major advantage of the team based approach is less space utilization requirement. Bukchin & Masin (2004) propose a multi-objective design of team oriented assembly system. This approach requires the bill of materials (BOM) for assembly. The BOM for product differs from the BOM for assembly. In a highly dynamic environment, it is very difficult to update both of their BOMs for product and assembly. Therefore the weakness of the proposed approach is that it requires the synchronization of a huge data for assembly system design. In this paper, an assembly system, in which both teamwork and mixed model processing are considered, has been studied. The reason for the modeling approach in this research study is the problems existed by nature in a real life assembly system. This assembly system, which is using teams and mixed model assembly, is very complicated, on which over 1500 assembly tasks are performed and nearly 300 models of tractors are covered. Regarding the complexity, many parameters relating to the design and management of the system can not be controlled by the engineers. The parameters are required to use for the system performance improvements. To solve these problems, a project team was built, and a project management framework was developed. The project team has studied and worked according to that project management framework. At the first stage of the project, the related books and papers were examined and discussed. After the comprehensive studies had been performed, a methodology for assembly system design and management was developed. The methodology consists of five steps as shown in Figure 1. The algorithms for each step are also coded using Matlab, and MS Excel is used as user interface.