How Covid-19 improved supermarket lead time and on-time delivery

This post is adapted from something I posted on LinkedIn a few weeks back (you can find the original here).

The way you want to optimize your production, increasing utilization of your bottleneck, improving flows and reducing your lead time can be found in supermarkets today. Probably without even knowing it, supermarkets took three major steps in applying the theory of constraints created by E. Goldratt and the good news is that it can be applied in your production, let’s see how!

Supermarkets are known to be a great example of cash management: customers pay them before they pay their suppliers, the dream of many manufacturers! We cannot apply this to the manufacturing world, but what we can learn today from supermarkets with the Covid-19 virus is how to reduce manufacturing lead time and improve on-time delivery.

Supermarkets were forced to apply Little’s law from queuing theory which states that the long-term average number L of customers in a stationary system is equal to the long-term average effective arrival rate λ multiplied by the average time W that a customer spends in the system. Expressed algebraically the law is

L = λW


In a nutshell, lead time is a function of WIP and throughput. The higher the WIP the higher the lead time. The lower the throughput, the higher the lead time.

The three step change that we see in supermarkets is:


Shopping queue
Photo by John Cameron on Unsplash

Before everyone was on lockdown supermarkets saw an enormous increase in demand. Everyone was stocking up for war. In order to cope with the additional demand, they increased throughput at their bottleneck – check-out – using temporary workers and flexibility.

One cashier told me “among all the cashiers you see right now there is my boss and her boss as well as people usually filling up the shelves”. We would all have done it, but can you do it in your production? Do you know your level of flexibility?


Since confinement and mandatory social distancing, supermarkets were forced to put new measures in place. The supermarket next to my house cannot have more than 100 customers in the premises at the same time. To implement this they simply started to count input of customers and stopped as soon as they reached 100. In production terms it means working with a constant WIP of 100 manufacturing orders. In order to keep their WIP constant they count output and as soon as 10 people are out 10 people are allowed in. It is an application of the drum-buffer-rope principle from Goldratt’s Theory Of Constraints.

Drum-buffer-rope, you said?

This is an illustration of one of the most powerful methods he created.

The drum represents your bottleneck. By definition the output of your production cannot be higher than your bottleneck. It means that producing faster than the bottleneck on other equipment will just generate WIP without increasing output. Consequently, the production pace should be given by the bottleneck, hence the drum.

If you strictly follow Little’s law you would try to get the lowest WIP possible. Unfortunately, unpredictable events happen in production all the time. In order to prevent your bottleneck from stopping due to WIP being too low, you need to have a buffer.

The rope is an illustration of the bottleneck pulling the first steps of production. Pulling here means that the release of orders is ordered by the bottleneck. Once an order is finished on the bottleneck it pulls a new one into production.

In our example of the supermarket, check-out sets the pace (drum) and pulls release of customers in the supermarket (rope), keeping 100 customers in the supermarket as buffer.


As mentioned before, the bottleneck is the check-out. Protecting people is of course the most important thing during this crisis and this is what supermarkets have done, protecting at the same time their bottleneck. I was happily surprised to see that supermarkets put a lot of effort into this. I have seen very quickly at the supermarket close to my home, cashiers wearing protective masks and gloves, and each having a dispenser of hydro-alcoholic gel. On top of that, companies were fast in providing transparent plastic sheets to put in front of cashiers and protect them from the germs that can be brought in by customers.


Just a few questions you need to ask yourself:

  • How do I release orders in production?
  • How do I manage WIP?
  • How do I protect my bottleneck?

You may argue that people are waiting to enter the supermarket. Yes, they do, but they don’t at the check-out. If you take your workshop, the Theory Of Constraints tells us that if you push all orders in production, WIP will increase and lead time will increase and become unpredictable. If you have low WIP, you will get shorter and much more predictable lead time. So yes, your backlog will be higher and that can feel scary, but the good news is that you will be able to promise a delivery date to your customers that you are sure to keep.

If you have questions, please put them in comments and let’s discuss!

Stay safe and stay home.