The Principle of Flow and Pull
by Chris Butterworth of S A Partners
Too often the principle of flow and pull is seen as applicable only in a manufacturing environment. In fact, the principle can be applied in many different ways. On a recent trip to Kenya I had the pleasure of experiencing the "Matatu" bus service in Nairobi. It really got me thinking about the principle of flow and pull.
The system is very different from my hometown in Sydney, which has a very similar population size to Niarobi. In Sydney, buses are on a fixed timetable, readily available to look up on various travel apps and published at each bus stop. Buses are frequent and you can even track your bus’s real-time location on your phone. Like most big cities they are more frequent at peak hours, but despite this, are often full during these hours and relatively empty off peak.
So is this a good example of pull and flow? At one level, buses are pulled through at peak times to match customer demand based on historical demand patterns. Flow is dramatically improved through the use of dedicated bus lanes. Car drivers watch enviously from the queue of their stationary vehicles as buses shoot down the near empty bus lane. Occasionally frustration takes over better judgement, and a quick burst down the bus lane provides the thrill of movement and the often-illusionary progress as you are forced to wait to be allowed back into your lane a few hundred meters further on. If you are really lucky you also avoid the fine.
So there is some pull and certainly better flow for the buses than for the cars. Outside of peak times, the bus system is definitely more push than pull, with fixed timetables pushing out buses regardless of actual demand.
The system has recently gone cashless, which saves a huge amount of time in collection and counting of cash, and generally works very well. All you need to do is make sure you buy an electronic card and keep it charged up with virtual cash and all is well. That is unless you are a new visitor to the city and unfamiliar with the bus drivers’ inability to accept cash. I recently witnessed an incident with a tourist family trying to board a bus. English was not their first language, and the driver struggled to explain that just giving him more and more money would not work.
A couple of passengers tried to help – even by offering to use our cards, but you can only use them once on each journey. So after a lot of shouting and gesticulating from the driver, they eventually gave up and will hopefully be able to have a great laugh about the whole experience once they get home—which will hopefully not be by bus. So the electronic payment is intended to improve flow, and generally does, but it was pushed on to the customers, many of whom would still prefer to use cash.
Now back to Nairobi. There are no timetables, few marked bus stops out of the main central boarding points, and you struggle to track your bus’s location on your phone. The buses are small and take about 20 people - sitting down. I boarded a bus and took a seat. Luckily I knew it was the right bus because I had a guide. Also there was a very helpful man with a sign who stood on the pavement saying where the bus was going and encouraging passersby to catch it.
After five minutes, I wondered what time we would set off. My helpful guide shrugged his shoulders.
"It depends," he said. I decided to practice the five whys and get to the root cause of why we had been waiting. “Well,” says my guide, "the bus is not full."
Eventually my root-cause analysis revealed we would not leave until every seat was occupied. Now other passengers started to join in and promote the virtues of the bus by shouting enthusiastically out the windows to anyone showing the slightest interest. We quickly had every seat full and were off.
A conductor came to collect our fare and asked where we wanted to be dropped off. We picked a well-known building on the bus route a short walk from our hotel and paid a fraction of a Sydney bus fare. It was cash only and calculated based on how far we traveled. The atmosphere on the bus was fantastic. It was a tight fit, and it was impossible not to get to know your fellow passengers. Sydney buses tend to be silent with people buried in their phones and not knowing where to look. We had a great laugh on the Matutu with everyone sharing a smile and enjoying the experience.
So is the Matatu a push or a pull system? In one respect it's definitely batch and queue with customers waiting for the full bus or "batch" to be completed before the journey can start. My initial reaction was that this is not a lean system but on reflection I realized it has its advantages.
- Its top customer value is cost and it certainly delivers on this by only travelling when full utilization is very high and cost per journey is very low.
- Fewer buses are needed in the fleet, saving on capital and maintenance costs.
- At peak times the wait time is very short as buses fill quickly.
- In off-peak times there are fewer buses driving around empty and are only pulled through based on demand.
So whilst the departure and arrival time may be a little more uncertain than the Sydney system, overall the Matatu provides a great solution that minimizes cost and provides customer flow with on-demand pull. "Kenyan time" has a bit more flexibility than Sydney time, but then it only takes one accident in Sydney for the whole timetable to get disrupted.
We need to ensure that we design our flow and pull systems to maximize customer value. Different customers value different things, and we must be careful not to try to impose one-size-fits-all in our systems design. Both the Matatu system and Sydney system work well in different contexts and both have elements of flow and pull.
Join the Shingo Institute mailing list and get more great articles!