For years, managers at Procter & Gamble, the consumer products family, told warehouse workers to pack a truck full of detergent and send it out to make deliveries to retailers, then pack the next truck with toothpaste, and so on. Then in the early 2000s, the company invested in inventory-tracking software that allowed managers to determine how many workers and how much time were required to get products to their ultimate destinations. What they discovered at first seemed counterintuitive. Sending out trucks fully loaded with a single product required the use of more inputs per unit of time than shipping a mix of products on trucks that were less full.

Packing trucks as fully as possible, the data revealed, left warehouse workers idle for long stretches between trucks. Also, filling individual trucks with boxes of a single product to be dropped off at many locations ultimately required driving more miles and using more fuel than loading a variety of products onto trucks driven to fewer final destinations. Based on this information, Procter & Gamble began sending out rerouted and less full trucks carrying more than one product at a time. The company was able to ship 30 percent more output per unit of driving time using the same truck, labor, and fuel inputs as before.

If Procter & Gamble continued to manufacture and ship as much output each week after the change in its distribution system, did it require the same amount of trucks, fuel, and workers to accomplish this task?