Real supply chain networks have high levels of operational variability. It takes a different amount of time to complete each subcycle from one pass the next. People vary in how long they take to make decisions and perform their work. Logistics varies in the amount of transit time it takes to move freight and in the time it takes to clear customs. Cash payments vary in how long it takes for them to be processed, and information access time varies in how long it takes to connect to the system to transfer the right information files. It is useful to think of loop velocity in terms of a probability distribution function. The measure of velocity varies a little each time around the loop. Rather than saying the loop velocity will be exactly X, it should be said that the loop velocity will probably fall somewhere between the bounds of Y ₁ to Y ₂.

The arcs of material flow, information flow, and cash flow that connect the trading partners have an important set of characteristics. Each arc connects a point of origin with a point of destination. Each arc has a cost and a mean time associated with it. Each arc also has a time variability. The threshold of network competitiveness is improved by maximizing velocity while minimizing variability. The days and weeks of queue times, manufacturing and distribution cycle times, logistics transit times, customs clearance times, and time driven by management policy decisions...