When Value-Stream Mapping Becomes Inadequate - Variability
Author: Descreye Solutions
The is article 1 of a series of articles on improving value-stream mapping. Value-stream mapping (VSM) is a valuable tool for improving systems. However, many people start using value-stream mapping without first recognizing the inherent limitations in the tool. They then don't see the expected result when they finish their improvement projects.
The biggest assumption that is made when making value-stream maps is that the system lacks variability. A value-stream map is created by using average cycle times, average customer demand, and average inventory buffer wait time. However, rarely is the assumption of no variability even close to the reality. This drives poor decision making, because if a system is analyzed based on its average cycle times it will always look underutilized and more capable than it ever is in reality. Often one can see this in VSM's where all the processes are less than 100% utilized, but there are many days worth of inventory in the process. Many VSM enthusiasts encourage to work towards 80% utilization in hopes that adding a utilization buffer will be enough to deal with the variability of the process. In reality traditional VSM is not built to deal with system variability.
Simulation is built for analyzing a systems variability. By using the variability to model the system dynamics a simulation accurately reflects the real-life scenario. Most VSM enthusiasts would not argue that a simulation model can be a more accurate reflection of the process. However, they would argue that the simplicity of a value-stream map and the complexity of creating simulations makes the accuracy argument less relevant in a debate of the two. In order to overcome this complexity Descreye Solutions has created OPS. OPS is a simple tool that takes less time to create a simulation than it would take to draw a VSM. This makes it both accurate and useful.
While VSM is a very useful tool for process improvement, it should be well understood before being used for system analysis. The issue of variability must be considered in complex systems, and is not adequately analyzed when using a traditional VSM. As such, when one uses VSM it should be with an underlying knowledge of the variation that is inherent in the system being analyzed.
← Back to blog