UC Berkeley

Production systems subject to variability always suffer a degradation in performance as variability will get absorbed by three kinds of buffers, inventory-, time-, and capacity buffers, in some combination. If buffers are not intentionally designed-in, variability will likely result in inventory build-up and time delays. However, the size and location of these three kinds of buffers can be decided in the process of designing a production system to achieve certain performance objectives. This paper reports on a study that focuses on balancing the use of inventory-and-time versus capacity as potential buffers, specifically in project production systems with various degrees of variability. The system design question is whether to carry buffers as inventory-time or as capacity while considering how their combination affects project duration, trade time-on-site, trade productivity, and inventory. To address possible answers to this question, we use discrete-event simulation to study variants of the Parade of Trades, a well-established ‘serious game’ that illustrates how variability in handoffs impacts the performance of a project comprised of a linear sequence of five trades. This simplicity of this game helps to develop managerial intuition about system performance. Inventory-time buffering absorbs variability with time and allows for higher resource utilization but is likely to extend the project duration. In contrast, capacity buffering absorbs variability by providing ‘extra’ production resources but may leave those resources underutilized. This tradeoff is of relevance to practitioners who design project production systems, including those involved in Lean Construction, Advanced Work Packaging (AWP), and AWP+.