Balancing a U-Shaped Assembly Line by Applying Nested Partitions Method
- Iowa State Univ., Ames, IA (United States)
In this study, we applied the Nested Partitions method to a U-line balancing problem and conducted experiments to evaluate the application. From the results, it is quite evident that the Nested Partitions method provided near optimal solutions (optimal in some cases). Besides, the execution time is quite short as compared to the Branch and Bound algorithm. However, for larger data sets, the algorithm took significantly longer times for execution. One of the reasons could be the way in which the random samples are generated. In the present study, a random sample is a solution in itself which requires assignment of tasks to various stations. The time taken to assign tasks to stations is directly proportional to the number of tasks. Thus, if the number of tasks increases, the time taken to generate random samples for the different regions also increases. The performance index for the Nested Partitions method in the present study was the number of stations in the random solutions (samples) generated. The total idle time for the samples can be used as another performance index. ULINO method is known to have used a combination of bounds to come up with good solutions. This approach of combining different performance indices can be used to evaluate the random samples and obtain even better solutions. Here, we used deterministic time values for the tasks. In industries where majority of tasks are performed manually, the stochastic version of the problem could be of vital importance. Experimenting with different objective functions (No. of stations was used in this study) could be of some significance to some industries where in the cost associated with creation of a new station is not the same. For such industries, the results obtained by using the present approach will not be of much value. Labor costs, task incompletion costs or a combination of those can be effectively used as alternate objective functions.
- Research Organization:
- Ames Lab., Ames, IA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- W-7405-Eng-82
- OSTI ID:
- 861605
- Report Number(s):
- IS-T 2504; TRN: US200602%%210
- Country of Publication:
- United States
- Language:
- English
Similar Records
A Multistage Stochastic Transmission Expansion Algorithm for Wide-Area Planning under Uncertainty
Electric Grid Expansion Planning with High Levels of Variable Generation