Browsing by Author "Pekelli, Sinem"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Conference Object
    A Multi-sided and Multi-model Assembly Line Balancing Problem
    (Springer Science and Business Media Deutschland GmbH, 2021) Seda Gemici; Emine Otuzbir; İrem Almila Koçyiğit; Sinem Pekelli; Fethi Tüzmen; Hande Oztop; Levent Kandiller; Pekelli, Sinem; Otuzbir, Emine; Gemici, Seda; Koçyiğit, İrem Almila; Tüzmen, Fethi; Öztop, Hande; Kandiller, Levent; N.M. Durakbasa , M.G. Gençyılmaz
    In this paper we study a real-life assembly line balancing problem (ALBP) of a cooler manufacturer brand in Manisa Turkey. The aim of this study is to create an effective assembly line balancing tool for the company which minimizes the number of stations and balances the total workloads of the stations while keeping the number of products produced the same. The studied ALBP is Type-1 ALBP that minimizes the number of workstations given a cycle time which is determined by a bottleneck operation. However different from the standard Type-1 ALBP some of the stations are two-sided stations in the studied assembly line. There are special constraints in the studied ALBP such as concurrent tasks preemptive tasks zone-restricted tasks and parallel tasks. Due to these additional characteristics of the system a novel mixed-integer linear programming (MILP) model is proposed for the studied ALBP to minimize the number of workstations. A secondary objective which balances the workload of the stations is also considered in the proposed MILP model using a lexicographic optimization. The computational experiments show that the proposed MILP model can obtain the optimal solution in reasonable computational time. When the model results are compared with the current system there is a 44% improvement in the number of stations on average. Furthermore a sensitivity analysis is performed to analyze the trade-off between the number of stations and cycle time criteria employing an ε-constraint method. Finally a user-friendly DSS is developed by embedding the proposed MILP model. © 2020 Elsevier B.V. All rights reserved.