Skip to main content

LeanOperationsService Improvement

Vaccine Lean

29 December 2021
Pic from
Pic from

I was absolutely delighted to receive my 3rd ‘booster’ vaccine a couple of weeks ago. Having been tremendously grateful for 2 x AstraZenecas, I have to confess that for my third Covid jab I was hoping for an RNA vaccine, preferably Pfizer, given the amount of gloating that I had had to suffer at the hands of my Pfizered Father.

When I rocked up to the Vaccination Centre, Moderna was the dish of the day. ‘Did you know that there’s an RNA hiding in the ModeRNA? No I didn’t?! Amazing!”. Whilst entertained at this new information I also sighed inside knowing that it would seem that Pfizer would again escape me, dooming me forever more to Dad teasing. Some quick googling in the vaccine queue allayed my fears that the Moderna vaccine possessed equal esteem however, so I chilled, awaiting my dose of magical RNA.

I say chilled, but, dear reader, what does a lean specialist do whilst in a (albeit, life-saving) mega queue? Why, one muses upon how the throughput could be improved!

Now, this blog comes with A NUMBER of health warnings and provisos namely:

  1. I have no intimate knowledge of the vaccination value stream other than being a patient within it. There might be all manner of things that are important within that I don’t know about. I apologise if this is the case and am really sorry for any factual inaccuracies or any suggestions I make that simply aren’t workable. Please let me know what they are and I’ll amend my thoughts accordingly.
  2. I am DEEPLY grateful for the speed and efficacy of NHS Wales in administering, not one, but THREE vaccines to me within 9 months, whilst in the middle of a global pandemic. The roll out has been phenomenal.
  3. I offer this blog not as ANY kind of preachy ‘letter of complaint’ (which I am known to be partial to). The (sane) adult population of Wales is going to be triple boosted before we know it! I offer it merely as an interesting exploration exercise into the variety of possibilities that operations management affords.

So whilst bearing these things in mind, let me begin.

One of the most important tools of an improvement specialist’s toolbox is to ‘go and see’ (genchi gembutsu or gemba gembutsu) and this is what I immersed myself in during my 45 minute wait.

Very quickly, I saw that the people who were doing the vaccinations were having to move around too much. When their last patient was finished, they moved to the waiting room space to collect their next patient. As well as this ‘motion’ (one of the lean wastes – employees having to move around) they were also having to do a lot of work on the computer.  The Theory of Constraints (Eli Goldratt) teaches us that critical parts of any process (sometimes bottleneck constraints, sometimes the urgent task that needs to be achieved) should be “free from constraints” and yes, the vaccinators were doing a lot more than injecting vaccines.

Another of my concerns was the amount of people, completely unconnected to each other, sharing an indoor space. The people closest to me were what the French apparently call ‘le manslipping‘ i.e. people that wear their masks under their nose (although to be fair to the male population there were les femmeslippers as well).   I felt deeply uncomfortable hanging out in a room with unmasked randoms for an hour. Another thing that I couldn’t grasp was the need for the initial ‘admission’ step.  Surely we should be just vaccinating everyone that attends at this point? Surely we no longer need the ‘are you able to have this vaccine?’ question and are able to just skip to the ‘let’s get some Covid RNA in your veins’ point, particularly when the vaccinators were looking you up and checking vaccine status as part of the cubicled vaccination process? To me this felt like a duplication of effort.

So, from memory, this was the situation in the vaccine centre that I experienced:

From memory map of the current vaccination process

What I haven’t depicted here is the waiting area post vaccine. It was sparsely populated. This might be because people weren’t listening to the ‘wait for 15 minutes to see if you have an adverse reaction’ advice and were simply walking out of the building, but I think it was more indicative of the pace of the vaccination process itself. The fact that there weren’t many people post vaccine who were waiting was indicative of the pace of the actual vaccination flow.

So let’s look at the value stream (the process flow) from MY perspective.

Not bad at all for a free, life saving vaccine. But the lean nut in me, and the thing that I teach in all of my executive education classes is ‘but let’s look at the difference between the end-to-end lead time and the value add’ and to speak the brutal truth, in this process, the value add is the vaccine going in… 2 seconds. No medical history, no chit chat to make you feel comfortable, no looking up patient records, no inputting into a computer. Jab and push. Done.

But achieving a 2 seconds drumbeat throughput is COMPLETELY unrealistic and to be avoided at all costs because:

  1. It’s so important to check whether the patient is eligible and able to have the vaccine. Do they feel ok? Are they allergic to anything?
  2. Humans don’t tend to like needles that much so might not take that kindly to attacking them with one at a furious pace!
  3. Having patients sit and wait afterward vaccination to check that they haven’t had a bad reaction can definitely be seen as value. Plus…
  4. Do we even have the supplies to deliver vaccines every 2 seconds?!

So let’s re-adjust our sights. If we look at the patient value stream, an experience of 17 mins 30 seconds as a bare minimum is far more realistic a goal to strive towards where 2 minutes is spent looking up the patient, finding their record, talking to them about their allergies and health and then administering the vaccine and checking if they are ok.

As a process drumbeat, 17 mins 30 seconds per person is still quite a lot to stomach for a pure 2 second value add process in a global pandemic, however we are helped here because it’s fair to say that the 15 minutes post vaccine wait can be seen as an ‘offline’ part of the process. SMED or Single Minute Exchange of Die is a very useful lean tool which shows us how we can protect a core process, filling it as full of value adding activity as possible and taking as much ‘non critical’ value adding activity away from the core process. The best way to think about this, as I’ve described previously here, is to think of Formula 1 pitstops.  The pitstop is pure ‘value add’, change tyres, fit new part …. all of the supporting valuable parts of the process take place outside of the time critical activity E.g. the tyres are pre-heated, the equipment is engineered to be quick releasing and quick fitting, the pit team repeatedly practice to enable swift changeovers etc. So the wait after the vaccine can be seen as something that should take place outside of the core value adding activity.

So if we were to get really serious about maximising vaccination throughput, we’d be looking to work to:

And I’d suggest that arguably, these two activities could take place at the same time if two people were present when interacting with the patient.

So I’ve had a go at redesigning the flow. Critical differences are:

  • To expand the capacity of the critical activity, less administrative focus is placed on the vaccinator. Instead, each room is equipped with one person who calls up patient records and checks vaccine eligibility and another person who administers the vaccine. The vaccinator can confirm medical history with the patient (the administrator can check at the same time) and then the vaccine is administered. We’d need to conduct a few experiments to see how this process would be effectively divided between two people  the best.
  • Using the same number of staff and volunteers (20), this new flow effectively reduces the number of vaccination spaces from 15 cubicles to 9, but because of the reduction in transportation and employee motion, reducing the amount of lead time spent in administration, vaccination productivity can be dramatically increased.
  • The treatment happens not in an enclosed cubicle, but within a tunnel of activity, encouraging flow and speed and enabling Covid compliance. You’d probably need a small privacy screen to make people feel slightly more comfortable.
  • Rather than one mega queue, we move to a more dynamic multiple queue system. Each vaccination team has more visibility and control of their own queue and critically the person tasked with feeding the different queues with new appointments knows to hold back the queue forcing them to wait outside, feeding the queue via an empty ‘kanban’ square.

I suggest that one person is tasked to regularly replenish supplies within each vaccination tunnel – in lean this function is termed the ‘water spider’, someone who takes on the ‘waste’ of motion to protect the process with as much value-add activity as possible.

The major problem with this set up of course is where to put the people, post vaccine, who have to wait for 15 minutes to check for any adverse reactions. If you worked to a realistic 3 minute pulse rate (2 mins core, 1 min buffer for transport, chat, etc) that would mean that you’d need three chairs post vaccine tunnel exit to house those people who we were keeping an eye on (15 minutes = 5 vaccinations). So the flow would look more like this:

Which doesn’t overcome the ‘lots of people in the same indoor space’ dilemma. Other problems with this design is that the vaccinators might greatly need the relief of getting up and stretching the legs to collect the next patient.  My design would very much keep them purely to task,  a machine as part of a factory production line, although feasibly, the administrator and vaccinator could be trained to take turns, alleviating some of the monotony?! Also of course, it isn’t perhaps ideal for two people to spend a lot of time in close proximity given Covid….

Whilst writing this blog in between Quality Streets, Mince Pies and Baileys, I have learned of a welcomed update to the administering of RNA vaccines, a necessary change as Public Health Wales describes, to account for the need to increase vaccination rates.  No longer are they recommending the 15 minute post vaccine wait, meaning that the following set up is much more possible. This makes complete sense to me once you have spent a few hours thinking about how to increase vaccine productivity.

The advantage of this set up is that it should mean that more vaccinations are achieved each hour and, I think critically, it means that there are less people spending less time in an enclosed inside space.  The last thing that anyone wants is to contract Covid whilst being vaccinated against it!  (Noting that I have ABSOLUTELY NO IDEA whether this has ever happened! I am just seeking to avoid the discomfort I felt whilst being vaccinated).

Of course, my suggestion would need to be trialled and as I’ve said in my introduction, there could be major problems with it, but considering different options is always an interesting thing to do.

So anyway, it was a joy to have a bit more time to write a blog than usual, apologies for its length!  I’m usually trying to write one in a vaccine tunnel of efficiency. 😉