Posts Tagged ‘Systems’

If you put an ear to someones chest you can hear their heart “lub-dub lub-dub lub-dub”. The sound is caused by the valves in the heart closing, like softly slamming doors, as part of the wonderfully orchestrated process of pumping blood around the lungs and body. The heart is an impressive example of bioengineering but it was not designed – it evolved over time – its elegance and efficiency emerged over a long journey of emergent evolution.  The lub-dub is a comforting sound – it signals regularity, predictability, and stabilty; and was probably the first and most familiar sound each of heard in the womb. Our hearts are sensitive to our emotional state – and it is no accident that the beat of music mirrors the beat of the heart: slow means relaxed and fast means aroused.

Systems and processes have a heart beat too – but it is not usually audible. It can been seen though if the measures of a process are plotted as time series charts. Only artificial systems show constant and unwavering behaviour – rigidity –  natural systems have cycles.  The charts from natural systems show the “vital signs” of the system.  One chart tells us something of value – several charts considered together tell us much more.

We can measure and display the electrical activity of the heart over time – it is called an electrocardiogram (ECG) -literally “electric-heart-picture”; we can measure and display the movement of muscles, valves and blood by beaming ultrasound at the heart – an echocardiogram; we can visualise the pressure of the blood over time – a plethysmocardiogram; and we can visualise the sound the heart makes – a phonocardiogram. When we display the various cardiograms on the same time scale one above the other we get a much better understanding of how the heart is behaving  as a system. And if we have learned what to expect to see with in a normal heart we can look for deviations from healthy behaviour and use those to help us diagnose the cause.  With experience the task of diagnosis becomes a simple, effective and efficient pattern matching exercise.

The same is true of systems and processes – plotting the system metrics as time-series charts and searching for the tell-tale patterns of process disease can be a simple, quick and accurate technique: when you have learned what a “healthy” process looks like and which patterns are caused by which process “diseases”.  This skill is gained through Operations Management training and lots of practice with the guidance of an experienced practitioner. Without this investment in developing knowlewdge and understanding there is a high risk of making a wrong diagnosis and instituting an ineffective or even dangerous treatment.  Confidence is good – competence is even better.

The objective of process diagnostics is to identify where and when the LUBs and HUBs appear are in the system: a LUB is a “low utilisation bottleneck” and a HUB is a “high utilisation bottleneck”.  Both restrict flow but they do it in different ways and therefore require different management. If we confuse a LUB for a HUB and choose the wrong treatent we can unintentionally make the process sicker – or even kill the system completely. The intention is OK but if we are not competent the implementation will not be OK.

Improvement Science rests on two foundations stones – Operations Management and Human Factors – and managers of any process or system need an understanding of both and to be able to apply their knowledge in practice with competence and confidence.  Just as a doctor needs to understand how the heart works and how to apply this knowledge in clinical practice. Both technical and emotional capability is needed – the Head and the Heart need each other.                          

Improvement Science is the knowledge and experience required to improve … but to improve what?

Improve safety, delivery, quality, and productivity?

Yes – ultimately – but they are the outputs. What has to be improved to achieve these improved outputs? That is a much more interesting question.

The simple answer is “flow”. But flow of what? That is an even better question!

Let us consider a real example. Suppose we want to improve the safety, quality, delivery and productivity of our healthcare system – which we do – what “flows” do we need to consider?

The flow of patients is the obvious one – the observable, tangible flow of people with health issues who arrive and leave healthcare facilities such as GP practices, outpatient departments, wards, theatres, accident units, nursing homes, chemists, etc.

What other flows?

Healthcare is a service with an intangible product that is produced and consumed at the same time – and in for those reasons it is very different from manufacturing. The interaction between the patients and the carers is where the value is added and this implies that “flow of carers” is critical too. Carers are people – no one had yet invented a machine that cares.

As soon as we have two flows that interact we have a new consideration – how do we ensure that they are coordinated so that they are able to interact at the same place, same time, in the right way and is the right amount?

The flows are linked – they are interdependent – we have a system of flows and we cannot just focus on one flow or ignore the inter-dependencies. OK, so far so good. What other flows do we need to consider?

Healthcare is a problem-solving process and it is reliant on data – so the flow of data is essential – some of this is clinical data and related to the practice of care, and some of it is operational data and related to the process of care. Data flow supports the patient and carer flows.

What else?

Solving problems has two stages – making decisions and taking actions – in healthcare the decision is called diagnosis and the action is called treatment. Both may involve the use of materials (e.g. consumables, paper, sheets, drugs, dressings, food, etc) and equipment (e.g. beds, CT scanners, instruments, waste bins etc). The provision of materials and equipment are flows that require data and people to support and coordinate as well.

So far we have flows of patients, people, data, materials and equipment and all the flows are interconnected. This is getting complicated!

Anything else?

The work has to be done in a suitable environment so the buildings and estate need to be provided. This may not seem like a flow but it is – it just has a longer time scale and is more jerky than the other flows – planning-building-using a new hospital has a time span of decades.

Are we finished yet? Is anything needed to support the these flows?

Yes – the flow that links them all is money. Money flowing in is called revenue and investment and money flowing out is called costs and dividends and so long as revenue equals or exceeds costs over the long term the system can function. Money is like energy – work only happens when it is flowing – and if the money doesn’t flow to the right part at the right time and in the right amount then the performance of the whole system can suffer – because all the parts and flows are interdependent.

So, we have Seven Flows – Patients, People, Data, Materials, Equipment, Estate and Money – and when considering any process or system improvement we must remain mindful of all Seven because they are interdependent.

And that is a challenge for us because our caveman brains are not designed to solve seven-dimensional time-dependent problems! We are OK with one dimension, struggle with two, really struggle with three and that is about it. We have to face the reality that we cannot do this in our heads – we need assistance – we need tools to help us handle the Seven Flows simultaneously.

Fortunately these tools exist – so we just need to learn how to use them – and that is what Improvement Science is all about.

I love history – not the dry boring history of learning lists of dates – the inspiring history of how leaps in understanding happen after decades of apparently fruitless search.  One of the patterns that stands out for me in recent history is how the growth of the human population has mirrored the changes in our understanding of the Universe.  This pattern struck me as curious – given that this has happened only in the last 10,000 years – and it cannot be genetic evolution because the timescale is to short. So what has fuelled this population growth? On further investigation I discovered that the population growth is exponential rather than linear – and very recent – within the last 1000 years.  Exponential growth is a characteristic feature of a system that has a positive feedback loop in it that is not balanced by an equal and opposite negative feedback loop. So, what is being fed back into the system that is creating this unbalanced behaviour? My conclusion so far is “collective improvement in understanding”.

However, exponential growth has a dark side – it is not sustainable. At some point a negative feedback loop will exert itself – and there are two extremes to how fast this can happen: gradual or sudden. Sudden negative feedback is a shock is the one to avoid because it is usually followed by a dramatic reversal of growth which if catastrophic enough is fatal to the system.  When it is less sudden and less severe it can lead into repeating cycles  of growth and decline – boom and bust – which is just a more painful path to the same end.  This somewhat disquieting conclusion led me to conduct the thought experiment that is illustrated by the diagram: If our growth is fuelled by our ability to learn, to use and to maintain our collective knowledge what changes in how we do this must have happened over the last 1000 years?  Biologically we are social animals and using our genetic inheritance we seem only able to maintain about 100 active relationships – which explains the natural size of family groups where face-to-face communication is paramount.  To support a stable group that is larger than 100 we must have developed learned behaviours and social structures. History tells us that we created communities by differentiating into specialised functions and to be stable these were cooperative rather than competitive and the natural multiplier seems to be about 100.  A community with more than 10,000 people is difficult to sustain with an ad hoc power structure with a powerful leader and we develop collective “rules” and a more democratic design – which fuels another 100 fold expansion to 1 million – the order of magnitide of a country or city. Multiply by 100 again and we get the size that is typical of a country and the social structures required to achieve stablity on this scale are different again – we needed to develop a way of actively seeking new knowledge, continuously re-writing the rule books, and industrialising our knowkedge. This has only happened over the last 300 years.  The next multipler takes us to Ten Billion – the order of magnitude of the current global population – and it is at this stage that  our current systems seem to be struggling again.

From this geometric perspective we appear to be approaching a natural human system barrier that our current knowledge management methods seem inadequate to dismantle – and if we press on in denial then we face the prospect of a sudden and catastrophic change – for the worse. Regression to a bygone age would have the same effect because those systems are not designed to suport the global economy.

So, what would have to change in the way we manage our collective knowledge that would avoid a Big Crunch and would steer us to a stable and sustainable future?

Africa is a fascinating place.  According to a documentary that I saw last year we are ALL descended from a small tribe who escaped from North East Africa about 90,000 years ago. Our DNA carries clues to the story of our journey and it shows that modern man (Africans, Europeans, Asians, Chinese, Japanese, Australians, Americans, Russians etc) – all come from a common stock. It is salutory to reflect how short this time scale is, how successful this tribe has been in replacing all the other branches of the human evolutionary tree, and how the genetic differences between colours and creeds are almost insignificant.  All the evolution that has happened in the last 90,000 years that has transformed the world and the way we live is learned behaviour. This means that, unlike our genes, it is possible to turn the clock backwards 90,000 years in just one generation. To avoid this we need to observe how the descendents of the original tribe learned to do many new things – forced by their new surroundings to adapt or perish.  This is essence of Improvement Science – changing context continuously creates new challenges – from which we can learn, adapt and flourish.

To someone born in rural England a mobile phone appears to be a relatively small step on a relentless technological evolution – to someone born in rural Africa it is a radical and world-changing paradigm shift – one that has already changed their lives.  In some parts of Africa money is now managed using mobile phones and this holds the promise of bypassing the endemic bureaucratic and corrupt practices that so often strangle the greens shoots of innovation and improvement. Information and communication is the lifeblood of improvement and to introduce a communication technology that is reliable, effective, and affordable into a vast potential for cultural innovation is rather like introducing a match to the touchpaper of a firework. Once the fuse has started to fizz there is no going back. The name given to this destabilising phenomenon is “disruptive innovation” and fortunately it can work for the good of all – so long as we steer it in a win-win-win direction. And that is a big challenge because our history suggests that we find exploitation easier than evolution and exploitation always leads to lose-lose-lose outcomes.

So while our global tribe may have learned enough to create a global phone system we still have much to learn about how to create a global social system.