12.3 The Drum-Buffer-Rope Method
Transitioning from the Five Focusing Steps, we now address the method for managing and controlling the flow within a system: the drum-buffer-rope method, devised by Goldratt. This method is particularly useful in environments where numerous tasks or components must be processed, such as in an insurance company dealing with a multitude of quotes and policies, or a manufacturing facility where raw materials are transformed into finished products. Here, the goal is to balance the flow by ensuring all resources are synchronized with the pace of the bottleneck.
The drum represents the rhythm dictated by the bottleneck’s capacity, which, in turn, sets the pace for the entire process. The buffer is strategically placed inventory ahead of the bottleneck, ensuring it is never starved for work. This is critical because any idle time at the bottleneck equates to lost time across the entire process. The rope is the communication link that connects the bottleneck to upstream resources, controlling their output to match the rate of the bottleneck, thus maintaining a stable level of buffer inventory.
This method incorporates the concepts of statistical fluctuations and dependent events. Statistical fluctuations acknowledge that there is inherent variability in the time each resource takes to complete a task. For instance, an executive performing a 3-way match to pay suppliers might take 20 minutes for a straightforward case, but several hours if discrepancies arise. When resources are sequentially dependent, these variations can lead to excess WIP or idle time.
Consider the two resources illustrated in the accompanying diagram. Resource 1 has a variable capacity, fluctuating between 8 to 12 pieces per minute, whereas Resource 2 consistently processes exactly 10 pieces per minute. At the commencement of operations, should Resource 1 produce merely 8 pieces, Resource 2 will be underutilized, awaiting the shortfall. In the following cycle, if Resource 1 outputs 12 pieces, Resource 2, constrained by its capacity, will process only 10, thus incrementing the Work in Process (WIP) upstream. This example typifies how statistical fluctuations can cause excessive WIP or idle time, underscoring the crucial need for a mechanism to balance the flow in response to variability, as previously discussed.
Even in this simplified scenario with two resources of equal average capacity—a theoretically balanced facility—we witness idle time and WIP accumulation due to statistical fluctuations and dependent events. These can give rise to ‘floating bottlenecks,’ where different resources become the constraint at various times because of random fluctuations. TOC advises against attempting to balance the capacity of each resource. Instead, it advocates for balancing the flow of the entire facility by aligning it with the pace set by the bottleneck. The drum-buffer-rope method is designed precisely to achieve this balance, synchronizing the facility’s flow with the throughput of the bottleneck.
