Controlling Flow under Uncertainty
One of the hallmarks of this book is the use of some unexpected sources for models of behavior. In this chapter, Reinertsen uses traffic control systems because they must adapt to constantly changing and uncertain conditions — accidents happen. Weather happens. Street repair happens. Some of these things happen on a schedule that can be anticipated and others cannot. Sound familiar?
What we learn from this example is that flow is a product of speed and density. Furthermore, small changes in either can have a profound effect on flow. When either goes up significantly, congestion quickly ensues. As we already know, congestion for product developers means bigger queues, higher capacity utilization, delays and higher costs. How to manage this?
One of the best examples of what Reinertsen calls “controlled occupancy” is familiar to Californians. It is what they refer to as the “ramp meter.”

Ramp meters control the volume of vehicles on the freeway and thereby increase flow without actually limiting access. We can use similar mechanisms to manage WIP level. Ideas such as cadence, forecasting and synchronization are discussed. As is the case throughout the book, concepts introduced elsewhere are interwoven. Capacity utilization is revisited with an eye towards determining the correct margin to leave available
to ensure higher flow rates.https://www.bptrends.com/principles-of-product-development-flow-second-generation-lean-product-development-by-donald-g-reinertsten/