EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
1380. Bi-objective based scalable balancing of a reconfigurable assembly line
Invited abstract in session TD-60: Resource constrained scheduling, stream Project Management and Scheduling.
Tuesday, 14:30-16:00Room: S09 (building: 101)
Authors (first author is the speaker)
| 1. | Siwar Arbi
|
| IMT Atlantique | |
| 2. | Audrey Cerqueus
|
| DAPI, IMT Atlantique | |
| 3. | Evgeny Gurevsky
|
| LS2N, University of Nantes | |
| 4. | Alexandre Dolgui
|
| LS2N, IMT Atlantique |
Abstract
Balancing an assembly line involves assigning a required set of tasks to a given set of workstations, subject to task precedence constraints and achieving some production goals. The two studied goals are antagonistic. The first seeks to maximize line productivity, which means minimizing the load on the busiest workstation. The second aims to maximize line robustness, expressed by a specific measure, and lies in the potential presence of tasks with uncertain processing time. A reconfigurable environment enables an already provided assembly line balancing (or LB) to adjust or improve its level of productivity or robustness if necessary, by allocating extra resources to workstations. As their number is limited, it is important to use them as efficiently as possible, and therefore to keep only such allocations, called efficient reconfigurations (or ER), which are non-dominated in the Pareto sense for two objectives studied. Different LB options provide different varieties of ER. In order to compare them, we introduce the so-called measure of scalability, which is computed into two steps. At first, for each set of ER sharing the same number of used extra resources, the hyper-volume of the image of this set in the objective space is calculated. Then, the sum of all these hyper-volumes is viewed as the scalability measure for the studied LB. To find an LB with the greatest value of its scalability measure, a simulated annealing is developed and tested on instances of different size.
Keywords
- Production and Inventory Systems
- Robust Optimization
Status: accepted
Back to the list of papers