Category Archives: COVID

Addressing COVID19 gaps between data, models and decision brings us back to hierarchical PID control, folks!

Posted on by
Reply

I’d like to comment about the way how careful we should be when we use data (even if it’s accurate at the source and at its processing steps!), when we build models to extract some dependencies between elements of the data, and ultimately when we make decisions.

Long time ago (approx. 40 years), when my father took over as head of control engineering department in St. Petersburg electrical engineering institute (LETI) from the previous head, Professor Alexander Vavilov, at their school they were excited by exploring the idea of evolutionary synthesis of control systems. One crucial part of this study was the development of theory of structural synthesis, where models of the system at each level of granularity had to be adequate to the criteria of optimal control. (By the way, graphs were essential in those models)

The basic idea was that depending on the level of granularity (or hierarchy) considered by the modeller, the system can have completely different criteria of correctness and/or optimality, hence certain aspects that are significant at small scale may not be important at a larger scale.
A bit like the criteria of control in the national level is not the same as criteria for control at the municipal level, and not the same at the level of local community, and not the same at the level of family units and individual households.
So, because of these differences and clashes of interest between different levels there is a lot of anxiety and misunderstanding in societies.

So, what the relationship with COVID-19?

Well the relationship is direct.

Let’s take the data on Mortality 2017 from the UK National Statistics: https://www.ons.gov.uk/visualisations/dvc509/chart1/index.html

This data shows that the number of deaths across the country in one year is significant – hundreds of thousands – not far from 1M. The relative number of deaths, that we witness now as a result of COVID-19 even if it will hit 10K-20K would be quite small though.

So, we clearly have different perspectives here, one is national (spatial) that stretches across the whole year (temporal), while the other could be local (e.g. an area of population in London) and taken during these 2-3 weeks of March-April. The relative increase in the number of deaths at the national scale is a small bump on the curve. I.e., integrating the number of deaths, caused by respiratory problems thanks to COVID-19, at the national scale will not give much effect to the game of the totals.

However, if we look spatiotemporally at the small scale we may see a significant rise in terms of differential and even proportional response. So, if we are particularly sensitive to these two aspects, differential and proportional, we may actually decide to react with a powerful action.

What we are facing here is exactly what I started with in my blog. We are facing with the different levels of granularity (or hierarchy, whatever we call it). Consider the coarse granularity. From this point of view our Mother Nature in us may say, well, why bother, the integral response (let’s denote it by letter I) is very small, and we look at time intervals of decades, so there is no need for any great change in decision-making. The problems of environmental nature are much more serious.

But let’s go down to the level of individuals, especially those living in the most affected areas of COVID-19. Again, our Mother Nature in us would tell us, that the rise in deaths due to coronavirus is an alarm, it may trigger a disaster, we may lose the loved ones, lose a job and income. What’s happening here? It’s actually that at the lower granularity level, the criteria for decision-making are based on differential and proportional responses (let’s denote them D and P). So, in mathematical terms at different levels of granularity we apply different coefficients, or what engineers call gains, to these aspects P, I and D, and form our decisions according to those gains or criteria of importance.

So, ultimately it is vital that the data we use, and the models which characterise this data in time and in space, where we calculate partial or full derivatives and integrate in space and time, or proportionalise in space and time, must be adequate to the criteria of significance we apply, and lead to corresponding decision-making at the appropriate level.

No doubt, the nations that are harmoniously hierarchical and fractally uniform, may have less problems in matching criteria of optimality with the P, I and D responses brought be the models from the actual data.

Yet, again we face that PID-control seems to rule the world we live in!

COVID-19 – Why China Did What it Did

Posted on by
Reply

From the horse’s mouth. Received this morning from a Chinese  source who is a top class engineering expert.

Very revealing!

Some of the actions of the Chinese government, which seemed counter-intuitive at the time, became quite clear from this explanation.

  1. How the hell did they decide to close up Wuhan when the official death figure was only 30 something? 
    Remember that the city is a uniquely important communications hub with air, rail and river transport crossing in multiple directions (in a war they’d probably prioritize bombing the place). The time was just before the Spring Festival before the annual spring travel crush started. Closing Wuhan spoils the SF(CNY) for a huge number of people, hurts the feelings of even more and damages the economy significantly. The modelling teams were assembled much earlier than this date and this action was significantly model-driven. The models tested different actions and the actual sequence was chosen as the least bad one. Closing Wuhan on its own looked stupid to some degree, but not as the first of the sequence of actions that followed:
  2. What about the rest of the country then?
    The rest of the country was allowed to continue through the first phase of the spring travel rush, which decanted probably 1/3 of the population from large cities onto the countryside, then the entire country was closed down preventing their return. This prevented the appearance of another Wuhan, with which the government would have no way of dealing.
  3. Volunteering albeit under peer pressure is a key
    As it happened, they were able to assemble large teams of medics from elsewhere in the country (the so-called volunteers – if you were a party member not volunteering was not an option, and non-members esp. low ranking nurses had incentives such as conversion from contract worker to full-time permanent worker) to descend on Wuhan and its province Hubei en masse. This depletion of medical strengths elsewhere proved sustainable because another flareup never happened. The President did not formally thank the people of Wuhan on behalf of the nation for nothing. When the people of China hear western media portray this as an apology for government errors they find this play quite difficult to imagine/understand. The hard/cold decision was to contain the spread locally from the first and therefore those local people had to suffer more hardships without volunteering. The least the nation could do is to appreciate this.
  4. Fangcang – makeshift hospitals are effective
    The establishment of the fangcang (makeshift hospitals using stadiums and exhibition centres) seemed strange, given that you were assembling ‘suspected cases’ all in one single space. The models predicted success which was borne out by reality. This has to do with how you want to deal with suspected cases and confirmed cases with light symptoms. It was determined that these people are better assembled together under professional care and control than remain at home to self-isolate with family. Fangcang-induced infections turned out to be negligible, almost zero. With beds a few metres from each other and everyone breathing the same air how was this possible? The answers are in the obligatory wearing of masks, on-hand medical and professional help and admin and enforced discipline, and almost continuous cleaning of the environment. These put together turned out to be vastly preferable, so far as the numbers are concerned, to home isolation where people do it any amateur manner they like/can.
  5. Testing methods with replication are crucial (real engineers can appreciate the use of time redundancy and diversity)
    The testing method adopted has practically 100% accuracy in the lab, close enough to 100% to be dependable for a tested population where the infection rate is only 1%, but in the field negative results were not trust-worthy (positives are completely fine). This was also put into the models and the resulting standard changes converted a large number of suspects to confirmed in a single day (all such converted cases had negative test results, but did not pass a CT scan test). The scientists read the UK’s confident reporting of how many tested with a large proportion of negatives with fascination, and speculate that the UK may have a more reliable testing procedure. This testing situation also inspired the fangcang approach as well as the very tight lockdown measures taken across the country. You don’t get cleared just because you had a negative. You need 2-3 negatives in a row without symptoms. In other words, treat everyone as a suspect case and everyone with symptoms as a confirmed case and design your control measures based on this assumption. The CCP is able to do this, other countries maybe not.
  6. Modelling approaches, also diverse and competing, are a must.
    The modelling gravitated towards two competing camps, by design of the government organizers. One is called the maths model and the other the medicine model. The first is led by system theorists and the second, epidemiologists. The commonly seen model of first order differential equation with an R0 factor is nowhere to be seen in either groups of models actually consulted by the decision makers – they are much more sophisticated than that. The maths model consistently returned more accurate predictions with worst case on death numbers error below 7% at all stages – this is the only hard number my friend was willing to disclose. All published models, either from within or without China which have appeared have been comparatively checked with the decision models and found to be inferior, usually by a lot.
  7. Future of the models?
    There is very little chance of seeing these decision models published, not any time soon. My friend’s words: “We should not publish when there is an atmosphere in which such a publication might result in extra-science interpretations and uses” and such an atmosphere will linger for a long time, by the looks of it. I read the CCP propaganda as well as the stuff coming out of our government and can see this stuff buried deep for long. However the modellers continue to work on data from the wider world now and the government continues to listen to them. One difference between China and much of the rest of the world is that the scientists cannot just tell the government the science says this and that without providing evidence, as the members of the government can understand scientific evidence at an academic level. And they organize multiple teams to work against each other to form a peer-review like environment from the start.
  8. Protection of medics is a key factor
    The most important issue, highlighted by the models and tested in real life, is the protection of the medics. Initially the disaster was when Wuhan people crowded general-purpose hospitals where the medics were not protected. When the external teams went to Wuhan+Hubei they were well prepared and formed special-purpose facilities which had a far greater success rate with next to zero infection of medics. Although this is intuitive, the actual numerical differences made in the deaths was unintuitively large.
  9. Ventilators is a last resort when it’s 20% survival chance left.
    One of the little-publicized facts is that the starting and ending procedures of ventilator use on a patient (putting them on/off the machine) represents the standing-out worst point for medic infections. This has caused a reluctance in China of using ventilators and the threshold for their use is set quite high, leading to ventilated patients having only a 20% rate of survival – if you are not already dying you are not ventilated. So they are a bit fascinated by the current western thing about seeing ventilators as some sort of almighty saviour, esp. given the current suboptimal PPE state for medics in an environment of retired medics (presumably not young) re-joining service.
  10. Masks, hand washing – NOT to be neglected 
    On how to protect ourselves, my friend emphasizes mask wearing and hand washing – diligent mask wearing and hand washing mimics the fangcang regime to some degree. Contrary to common belief, the wearing of even three-ply surgery masks protects not only the environment from the wearer but also the wearer from the environment, and N95 masks are indeed better. He became a bit rhetorical and urged us to disregard imagined stigmatization to prioritize life, both our own and that of those who may stigmatize us.