A few weeks ago we posted a video of how Toyota engineers helped a New York City food bank optimize how it packs boxes of food to be distributed to families still displaced from Hurricane Sandy. Now the New York Times has a more detailed story of how Toyota Production System Support Center has worked with various parts of The Food Bank for New York City (In Lieu of Money, Toyota Donates Efficiency to New York Charity, Jul 26). It turns out that packing boxes is only one project out of many.
One of the more interesting projects is speeding up how quickly clients can be seated for dinner at a Harlem soup kitchen.
Toyota’s engineers went to work. The kitchen, which can seat 50 people, typically opened for dinner at 4 p.m., and when all the chairs were filled, a line would form outside. Mr. Foriest would wait for enough space to open up to allow 10 people in. The average wait time could be up to an hour and a half.
Toyota made three changes. They eliminated the 10-at-a-time system, allowing diners to flow in one by one as soon as a chair was free. Next, a waiting area was set up inside where people lined up closer to where they would pick up food trays. Finally, an employee was assigned the sole duty of spotting empty seats so they could be filled quickly. The average wait time dropped to 18 minutes and more people were fed.
Why do I like this example? For one, it is pretty common service problem. Lots of services have to deal with very peaked demand patterns so waits become an issue. This is a nice example that shows the power of lean operations to address these issues in service settings.
Second, one of the points that we make when teaching lean operations in the core class is about reducing the batch size. We argue that lowering the batch size can reduce the amount of inventory in the system and thus the flow time. That is what is happening here. The old policy of only letting in ten at a time created an artificial batch. I suspect that eliminating that batching was a big reason for the drop in wait time. In particular, if dining room seats are the bottleneck of the process, batching arrivals means losing capacity. Eliminating batching minimizes how much capacity is wasted.
Finally, it is nice example of adding resources as a countermeasure. There is actually a reason to have some batching in admitting people. It would likely guarantee that once a client clears the serving line they find a seat relatively quickly. I suspect it is none too pleasant for all concerned to force a soup kitchen client to wait to be seated. If people aren’t admitted until there are several empty seats, most of the ten people entering have an obvious place to go once they are served. Now with clients moving through as a batch of one, spotting an empty seat becomes more of challenge. Hence, the need for a countermeasure — i.e., a dedicated set of eyes to help spot an empty chair. It is not even clear that this requires an extra body. With one person at a time moving through after the initial wave of clients, getting people through the service line doesn’t have to be too fast. That is, it may be possible to move a worker from slinging grub to monitoring tables without impacting service.