Allocation of work to the stations of an assembly line with buffers between stations and three general learning patterns. (14th January 2008)

• Record Type: Journal Article
• Authors: Cohen, Yuval; Dar-El, Ezey M.; Vitner, Gad; Sarin, Subhash C.
• Is Part Of: International journal of intelligent systems technologies and applications. Volume 4:Number 1/2(2008)
• Page Start: 123
• Page End: 140
• Publication Date: 2008-01-14
• Subjects: batch assembly -- learning -- work-in-process -- WIP -- assembly line balancing -- ALB -- optimisation -- work allocation -- buffer capacity -- makespan reduction; Artificial intelligence -- Periodicals; Intelligent control systems -- Periodicals; 006.3
• Languages: English
• ISSNs: 1740-8865

Abstract

This paper addresses the problem of allocating work to the stations of an assembly line for minimising the makespan required to process a lot of products with a low overall demand. This environment is characterised by different learning slopes in the various stations (due to the nature of work). We assume small (e.g. a laser pen) to medium size products (e.g. a pilot helmet) so buffer space, for temporary purposes, is typically not a problem. Three general patterns of stations' learning are considered, namely, decreasing, increasing and constant. Methodologies are presented for the optimal allocation of work to the stations for each of these cases. Some numerical results are also presented to show a significant impact of buffer capacity on makespan reduction as well as to reveal significant improvement in the makespan value due to unequal (but optimal) work allocation.