We consider a user who buys a number of identical capital goods systems (e.g., medical, manufacturing, or communication systems) for which she must have very high availability. In such a situation, there are typically several options that can be used to facilitate this availability. Often, the user can choose to build in cold standby redundancy for critical components. She may also typically buy spare parts with the systems so that during their exploitation phase, when a part in a system fails, the failed part can be replaced by
a ready-for-use part from inventory. In addition, an emergency procedure is usually available by which a part is shipped from a distant central warehouse (at an additional cost) to be applied when there is a stock out. To these options we introduce another: The possibility of initiating an emergency shipment when stock is one. Thus, the user may choose one of three policies per component: The different combinations of the redundancy decision and the timing of applications of the emergency procedure. (In addition, she must
decide how much spare parts inventory to purchase, for any policy.) Each policy provides different total uptime against different total costs incurred. We formulate the problem as the minimization of the total costs incurred for the systems over their lifetimes, under a constraint for the total uptime of all systems. These total costs consist of acquisition costs, spare parts costs, and repair costs. We optimally solve the problem by decomposing the multi-component problem into single-component problems, and then conducting exact analysis on these single-component problems, we derive results on when each of the three policies is optimal. Using these, we construct an efficient frontier which reflects the trade-off between the uptime and the total costs of the systems. In addition, we provide a method to rank the components by the relative value of investing in redundancy. We illustrate these results through numerical examples.
|Place of Publication||Eindhoven|
|Publisher||Technische Universiteit Eindhoven|
|Number of pages||34|
|Publication status||Published - 2011|
|Name||BETA publicatie : working papers|