Article ID: | iaor20022182 |
Country: | Netherlands |
Volume: | 41 |
Issue: | 4 |
Start Page Number: | 405 |
End Page Number: | 421 |
Publication Date: | Feb 2002 |
Journal: | Computers & Industrial Engineering |
Authors: | Dar-El Ezey M., Rubinovitz Jacob, Bukchin Joseph |
Keywords: | make-to-order |
Mixed model assembly lines can be found today in many industrial environments. With the growing trend for greater product variability and shorter life cycles, they are replacing the traditional mass production assembly lines. In many cases, these lines follow a ‘make-to-order’ production policy, which reduces the customer lead-time, and is expressed in a random arrival sequence of different model types to the line. Additional common characteristics of such mixed model lines in a make-to-order environment are: small numbers of work stations, a lack of mechanical conveyance, and highly skilled workers. The design problem of mixed model assembly lines in a make-to-order environment is addressed in this paper. A mathematical formulation is presented which considers the differences between our model and traditional models. A heuristic that minimizes the number of stations for a predetermined cycle time is developed consisting of three stages: the balancing of a combined precedence diagram, balancing each model type separately subject to the constraints resulting from the first stage, and a neighborhood search based improvement procedure.