Abstract
In this paper, we study the design of a transfer line consisting of two unreliable machines in series and an interstage buffer. In addition to the buffer capacity, the system planner can decide on the inventory levels for spare parts in order to reduce the time to repair failed machines. Our work aims to find cost-efficient combinations of buffer capacity and spare parts inventory. We formulate discrete-time Markov chains for two different situations in order to analyse the trade-off between buffer capacity, spare parts inventory and throughput. By means of numerical examples, we illustrate that the effect of a spare part on the efficiency of a transfer line is much greater than the effect of additional buffer places. We conclude that it is necessary to optimize buffer capacity and spare parts inventory levels simultaneously. For systems with identical critical components, our results show that not only is there a pooling effect of safety stocks but also between the buffer places and spare parts inventory levels. We conclude that planners who select a standardised component for the system design and optimize spare parts levels as well as buffer capacity can significantly reduce costs or increase the throughput of a transfer line.
Original language | English |
---|---|
Pages (from-to) | 121-134 |
Number of pages | 14 |
Journal | European Journal of Operational Research |
Volume | 284 |
Issue number | 1 |
DOIs | |
State | Published - 1 Jul 2020 |
Keywords
- Manufacturing systems
- Spare parts
- Stochastic models
- Transfer lines