Deconstructing Deming V – Improve constantly and forever

5. Improve constantly and forever the system of production and service, to improve quality and productivity, and thus constantly decrease costs.

W Edwards Deming Point 5 of Deming’s 14 Points. Surely about this there can be no controversy.

Improvement means reducing operating costs, enhancing customer value, and developing flexibility and agility. Improvement means constantly diminishing the misalignment between the Voice of the Process and the Voice of the Customer.

The UK awaits fresh productivity statistics next month but the figures up to the end of 2013 make sobering reading. UK productivity has been in miserable decline since 2008. In response to tightening of demand, failures of liquidity, absence of safe investment alternatives, rises in taxation and straightened cash flows, the aggregate response of industry has been a decline in human efficiency.

The reasons this has happened are no doubt complex. The paradox remains that it is improvement in productivity that grows sustainable rewards, captures markets and releases working capital for new ventures. At first sight it appears the answer to all the ills of a recession.

How will you know?

In their seminal model for improving productivity, Thomas Nolan and Lloyd Provost posed the question:

How will you know when a change is an improvement?

It is such a simple questions but it is too seldom asked and I suspect that itself is a major barrier to improvement.

We are beset by human induced change, by government and by business managers. The essential discipline is critically to question whether such change results in an improvement. It is an unpopular question. Nobody who champions a particular change wants to be proved wrong, or confronted with a marginal improvement that fails to live up to an extravagant promise.

Business measurement is mandated in the modern corporation. Businesses, governments, organisations abound with KPIs, metrics, “Big Ys”, results measures … and often a distracting argument over what to call them. There is no lack of numbers for answering the question. We are constantly assured that we now have the Big Data whose absence frustrated past strategy.

The habitual analytic tool in old-style businesses was what Don Wheeler mischievously named the executive time series, two numbers, one larger (or smaller) than the other, selected to show movement in the desired direction. That is, as Scottish folklorist Andrew Lang put it:

… using statistics in the same way that a drunk uses lamp-posts — for support rather than illumination.

It is a moral certainty that no two measurements will yield the same number. One will be larger than the other. It will be easy to select two to support any pet project or theory.

Building a persuasive case that improvement has happened firstly requires a rigorously constructed baseline. Without an objective description of the historical experience base, claims as to improvement are simply speculative.

And beyond that, what the executive time series cannot do is distinguish signal from noise. It cannot help because the answer to the question When will you know …? is When there is a signal in the data. That can only be answered with the diligent and rigorous use of process behaviour charts.

At the top of this page is a “RearView” tab. Without the trenchant and determined use of process behaviour charts there is not even a white line in the rear view mirror. The only signal will come from the “bang” when we hit the kerb.

What to improve

Deming’s further message was that it was every process that was to be improved, not simply those whose customer was the end consumer. Many processes have internal customers with their own voice. Processes of management of human resources, maintenance and accounting can all have a critical impact on organisation performance. They must keep on getting better too.

Being held to account is never comfortable but neither is the realisation that we have surrendered control of assets without the means of knowing when such assets are incrementally put to increasingly efficient, effective and agile use.

We need louder demands of “Show me!”

The dark side of discipline

W Edwards Deming was very impressed with Japanese railways. In Out of the Crisis (1986) he wrote this.

The economy of a single plan that will work is obvious. As an example, may I cite a proposed itinerary in Japan:

          1725 h Leave Taku City.
          1923 h Arrive Hakata.
Change trains.
          1924 h Leave Hakata [for Osaka, at 210 km/hr]

Only one minute to change trains? You don’t need a whole minute. You will have 30 seconds left over. No alternate plan was necessary.

My friend Bob King … while in Japan in November 1983 received these instructions to reach by train a company that he was to visit.

          0903 h Board the train. Pay no attention to trains at 0858, 0901.
          0957 h Off.

No further instruction was needed.

Deming seemed to assume that these outcomes were delivered by a capable and, moreover, stable system. That may well have been the case in 1983. However, by 2005 matters had drifted.

Aftermath of the Amagasaki rail crashThe other night I watched, recorded from the BBC, the documentary Brakeless: Why Trains Crash about the Amagasaki rail crash on 25 April 2005. I fear that it is no longer available in BBC iPlayer. However, most of the documentaries in this BBC Storyville strand are independently produced and usually have some limited theatrical release or are available elsewhere. I now see that the documentary is available here on Dailymotion.

The documentary painted a system of “discipline” on the railway where drivers were held directly responsible for outcomes, overridingly punctuality. This was not a documentary aimed at engineers but the first thing missing for me was any risk assessment of the way the railway was run. Perhaps it was there but it is difficult to see what thought process would lead to a failure to mitigate the risks of production pressures.

However, beyond that, for me the documentary raised some important issues of process discipline. We must be very careful when we make anyone working within a process responsible for its outputs. That sounds a strange thing to say but Paul Jennings at Rolls-Royce always used to remind me You can’t work on outcomes.

The difficulty that the Amagasaki train drivers had was that the railway was inherently subject to sources of variation over which the drivers had no control. In the face of those sources of variation, they were pressured to maintain the discipline of a punctual timetable. They way they did that was to transgress other dimensions of process discipline, in the Amagasaki case, speed limits.

Anybody at work must diligently follow the process given to them. But if that process does not deliver the intended outcome then that is the responsibility of the manager who owns the process, not the worker. When a worker, with the best of intentions, seeks independently to modify the process, they are in a poor position, constrained as they are by their own bounded rationality. They will inevitably by trapped by System 1 thinking.

Of course, it is great when workers can get involved with the manager’s efforts to align the voice of the process with the voice of the customer. However, the experimentation stops when they start operating the process live.

Fundamentally, it is a moral certainty that purblind pursuit of a target will lead to over-adjustment by the worker, what Deming called “tampering”. That in turn leads to increased costs, aggravated risk and vitiated consumer satisfaction.

Target and the Targeteers

This blog appeared on the Royal Statistical Society website Statslife on 29 May 2014

DartboardJohn Pullinger, newly appointed head of the UK Statistics Authority, has given a trenchant warning about the “unsophisticated” use of targets. As reported in The Times (London) (“Targets could be skewing the truth, statistics chief warns”, 26 May 2014 – paywall) he cautions:

Anywhere we have had targets, there is a danger that they become an end in themselves and people lose sight of what they’re trying to achieve. We have numbers everywhere but haven’t been well enough schooled on how to use them and that’s where problems occur.

He goes on.

The whole point of all these things is to change behaviour. The trick is to have a sophisticated understanding of what will happen when you put these things out.

Pullinger makes it clear that he is no opponent of targets, but that in the hands of the unskilled they can create perverse incentives, encouraging behaviour that distorts the system they sought to control and frustrating the very improvement they were implemented to achieve.

For example, two train companies are being assessed by the regulator for punctuality. A train is defined as “on-time” if it arrives within 5 minutes of schedule. The target is 95% punctuality.
TrainTargets
Evidently, simple management by target fails to reveal that Company 1 is doing better than Company 2 in offering a punctual service to its passengers. A simple statement of “95% punctuality (punctuality defined as arriving within 5 minutes of timetable)” discards much of the information in the data.

Further, when presented with a train that has slipped outside the 5 minute tolerance, a manager held solely to the target of 95% has no incentive to stop the late train from slipping even further behind. Certainly, if it puts further trains at risk of lateness, there will always be a temptation to strip it of all priority. Here, the target is not only a barrier to effective measurement and improvement, it is a threat to the proper operation of the railway. That is the point that Pullinger was seeking to make about the behaviour induced by the target.

And again, targets often provide only a “snapshot” rather than the “video” that discloses the information in the data that can be used for planning and managing an enterprise.

I am glad that Pullinger was not hesitant to remind users that proper deployment of system measurement requires an appreciation of psychology. Nobel Laureate psychologist Daniel Kahneman warns of the inherent human trait of thinking that What you see is all there is (WYSIATI). On their own, targets do little to guard against such bounded rationality.

In support of a corporate programme of improvement and integrated in a culture of rigorous data criticism, targets have manifest benefits. They communicate improvement priorities. They build confidence between interfacing processes. They provide constraints and parameters that prevent the system causing harm. Harm to others or harm to itself. What is important is that the targets do not become a shield to weak managers who wish to hide their lack of understanding of their own processes behind the defence that “all targets were met”.

However, all that requires some sophistication in approach. I think the following points provide a basis for auditing how an organisation is using targets.

Risk assessment

Targets should be risk assessed, anticipating realistic psychology and envisaging the range of behaviours the targets are likely to catalyse.

Customer focus

Anyone tasked with operating to a target should be periodically challenged with a review of the Voice of the Customer and how their own role contributes to the organisational system. The target is only an aid to the continual improvement of the alignment between the Voice of the Process and the Voice of the Customer. That is the only game in town.

Borrowed validation

Any organisation of any size will usually have independent data of sufficient borrowing strength to support mutual validation. There was a very good recent example of this in the UK where falling crime statistics, about which the public were rightly cynical and incredulous, were effectively validated by data collection from hospital emergency departments (Violent crime in England and Wales falls again, A&E data shows).

Over-adjustment

Mechanisms must be in place to deter over-adjustment, what W Edwards Deming called “tampering”, where naïve pursuit of a target adds variation and degrades performance.

Discipline

Employees must be left in no doubt that lack of care in maintaining the integrity of the organisational system and pursuing customer excellence will not be excused by mere adherence to a target, no matter how heroic.

Targets are for the guidance of the wise. To regard them as anything else is to ask them to do too much.

Ninety years on

Walter ShewhartOn 16 May 1924, ninety years ago today, Walter Shewhart sent his manager a short memo, no longer than one page. Shewhart described what came to be called the control chart, what we would today call a process behaviour chart.

Shewhart, a physicist by training and engineer by avocation, had been involved in improving the reliability of radio and telegraph hardware for the Western Electric Company. Equipment buried underground was often costly to repair and maintain. Shewhart had realised that the key to reliability improvement was reduction in manufactured product variation. If variation could, hypothetically, be eliminated then everything would work or everything would fail. If everything failed it would soon be fixed. Variability confounded improvement efforts.

Shewhart shared a profound insight about variation with a diverse group of independent contemporaries including Bruno de Finetti and W E Johnson. If we wanted to be able to reduce variation, we had to be able to predict it. Working to reduce variation turned on the ability to predict future behaviour.

De Finetti and Johnson were philosophers who didn’t go as far as turning their ideas into instrumental tools. The control chart turned out to be the silver bullet for predicting the future. It is a convivial tool. Shewhart invented it ninety years ago today.

If it isn’t supporting and guiding your predictions, you’re ninety years out of date (assuming that you’re reading today).

See the RearView tab at the top of the page for further background.

How to use data to promote your business

… or alternatively, five ways not to. Quite by chance, I recently came upon a paper by Neil H. Spencer and Lindsey Kevan de Lopez at the Statistical Services and Consultancy Unit of the University of Hertfordshire Business School entitled “Item-by-item sampling for promotional purposes”.

The abstract declares the aim of the paper.

In this paper we present a method for sampling items that are checked on a pass/fail basis, with a view to a claim being made about the success/failure rate for the purposes of promoting a company’s product/service.

The sort of statements the authors want to validate occur where all items outside some specification are classed as defective. I would hope that most organisations would want to protect the customer from defects like these but the authors of the paper seem to want to predicate their promotion on the defects’ escape. The statements are of the type:

There is a 95% probability that the true proportion of trains delayed by more than 5 minutes is less than 5%.

— or:

There is a 95% probability that the true proportion of widgets with diameter more than 1% from nominal is less than 5%.

I can see five reasons why you really shouldn’t try to promote your business with statements like this.

1. Telling your customers that your products are defective

Or to put it another way “Some of our products are defective. You might get one.” This might be a true statement at your current level of quality maturity but it is not something to shout at customers. All these statements do is to germinate doubt about your product in the customer’s mind. Customers want products and services that simply perform. Making customers think that they might not will be a turn off.

Customers will not think it okay to end up with a defective item or outcome. They will not put it down just to the “luck of the draw”. The products will come back but the customers won’t.

If you are lucky then your customer won’t even understand this type of promotional statement. There are just too many percentages. But they might remember the word “defect”.

2. Tolerating defects

Or to put it another way “Some of our products are defective and we don’t care.” Quoting the 5% defective with pride suggests that the producer thinks it okay to make and sell defects. In the 1980s Japanese motor manufacturers such as Toyota seized market share by producing reliable vehicles and using that as a basis for marketing and creating a reputation for quality.

Any competitive market is destined to go that way eventually. Paradoxically, what Toyota and others discovered is that the things you have to do to make things more reliable are the same things that reduce costs. Low price and high quality goods and services have an inbuilt advantage in penetrating markets.

3. Saying nothing about the product the customer is considering buying

The telling phrase is “true proportion of” trains/ widgets. As a matter of strict statistical technicality, Spencer and de Lopez don’t describe any “method for sampling” at all. They only describe a method of calculating sample size, worked out using Bayes’ theorem. Because they use the word “true”, it can only be that they were presuming what W Edwards Deming called an enumerative study, a characterisation of a particular sampling frame that yields information only about that frame. They took a particular slice of output and sampled that. Such a study is incapable of saying anything about future widgets or trains.

Put another way, “When we looked at a slice of our products we’re pretty sure that no more than 5% were defective. We don’t care. As to future products, we don’t know. Yours may be defective.”

I think we need a name for soi-disant Bayesians who chronically fail to address issues of exchangeability (stability and predictability).

4. Throwing away most of the valuable information on your product

Looking at the train example, “5% more than 5 minutes late” may mean:

  • “5% were 6 minutes late, the rest were 4 minutes late”; or
  • “4% were one hour late, 1% were cancelled and the rest were on time”; or

These various scenarios will have wholly different psychological and practical impacts on customers. Customers care which will happen to them.

Further, where we actually measure delay in minutes or diameter in millimetres, that is data that can be used to improve the business process and, with diligence, achieve the sort of excellence where quality failures and defects simply do not happen. That then provides the sort of consumer experience of satisfaction that can be developed into a public reputation for quality, performance and cost. That in turn supports promotional statements that will chime with customer aspirations and build business. Simply checking on a pass/ fail basis is an inadequate foundation for such improvement.

5. Managing by specifications

Taguchi1This is the subtlest point to turn your attention towards once everything is within specification. In the train example, the customer wants the train to be on time. Every deviation either side of that results in customer dissatisfaction. It also results in practical operating and timetable problems and costs for the railway. In the 1960s, Japanese statistician Genechi Taguchi put forward the idea that such losses should be recognised and form the basis of measuring improvement. The Taguchi loss function captures the idea that losses start with every departure from nominal and then start to escalate.

That leads to the practical insight that the improvement objective of any business process is “on target, minimum variation”.

What the Spencer-de Lopez statements ultimately say is that the vendor is willing to tolerate any train being 5 minutes late and 5% of trains being delayed even longer than that, perhaps indefinitely. Whether even that depressing standard is achieved in the future, who knows? Perhaps the customer will be lucky.

I fear that such statements will not promote your business. What will promote your business is using measurement to establish, maintain and improve process capability. That will provide the sort of excellent customer experience that can be mapped, promoted and fed back into confident, data based marketing campaigns aimed at enhancing reputation. Reputation supports talent recruitment and fosters a virtuous circle of excellence. This is what reputation management is about.

I do note that Spencer and de Lopez protest that this is only a working paper but it has been on their website since mid-2012 so I presume they are now owning the contents.

Just as a final caveat I think I should point out that the capability indices Cp and Cpk, though useful, do not measure Taguchi loss. That is the topic for another blog.

It was 20 years ago today …

W._Edwards_Deming[1]Today, 20 December 2013, marks the twentieth anniversary of the death of W Edwards Deming. Deming was a hugely influential figure in management science, in Japan during the 1950s, 1960s and 1970s, then internationally from the early 1980s until his death. His memory persists in a continuing debate about his thinking among a small and aging sector of the operational excellence community, and in a broader reputation as a “management guru”, one of the writers who from the 1980s onwards championed and popularised the causes of employee engagement and business growth through customer satisfaction.

Deming’s training had been in mathematics and physics but in his professional life he first developed into a statistician, largely because of the influence of Walter Shewhart, an early mentor. It was fundamental to Deming’s beliefs that an organisation could only be managed effectively with widespread business measurement and trenchant statistical criticism of data. In that way he anticipated writers of a later generation such as Nate Silver and Nassim Taleb.

Since Deming’s death the operational excellence landscape has become more densely populated. In particular, lean operations and Six Sigma have variously been seen as competitors for Deming’s approach, as successors, usurpers, as complementary, as development, or as tools or tool sets to be deployed within Deming’s business strategy. In many ways, the pragmatic development of lean and Six Sigma have exposed the discursive, anecdotal and sometimes gnomic way Deming liked to communicate. In his book Out of the Crisis: Quality, Productivity and Competitive Position (1982) minor points are expanded over whole chapters while major ideas are finessed in a few words. Having elevated the importance of measurement and a proper system for responding to data he goes on to observe that the most important numbers are unknown and unknowable. I fear that this has often been an obstacle to managers finding the hard science in Deming.

For me, the core of Deming’s thinking remains this. There is only one game in town, the continual improvement of the alignment between the voice of the process and the voice of the customer. That improvement is achieved by the diligent use of process behaviour charts. Pursuit of that aim will collaterally reduce organisational costs.

Deming pursued the idea further. He asked what kind of organisation could most effectively exploit process behaviour charts. He sought philosophical justifications for successful heuristics. It is here that his writing became more difficult to accept for many people. In his last book, The New Economics for Industry, Government, Education, he trespassed on broader issues usually reserved to politics and social science, areas in which he was poorly qualified to contribute. The problem with Deming’s later work is that where it is new, it is not economics, and where it is economics, it is not new. It is this part of his writing that has tended to attract a few persistent followers. What is sad about Deming’s continued following is the lack of challenge. Every seminal thinker’s works are subject to repeated criticism, re-evaluation and development. Not simply development by accumulation but development by revision, deletion and synthesis. It is here that Deming’s memory is badly served. At the top of the page is a link to Deming’s Wikipedia entry. It is disturbing that everything is stated as though a settled and triumphant truth, a treatment that contrasts with the fact that his work is now largely ignored in mainstream management. Managers have found in lean and Six Sigma systems they could implement, even if only partially. In Deming they have not.

What Deming deserves, now that a generation, a global telecommunications system and a world wide web separate us from him, is a robust criticism and challenge of his work. The statistical thinking at the heart is profound. For me, the question of what sort of organisation is best placed to exploit that thinking remains open. Now is the time for the re-evaluation because I believe that out of it we can join in reaching new levels of operational excellence.

Sad news on railway suicide statistics

I recently blogged about statistics of suicides on British railways here and here. Some very worthwhile programmes had been put in place with the objective of reducing these tragic deaths. However, my view at the point of my earlier posts was that this was a stable system of trouble, that there was neither a deteriorating trend nor any sign of improvement.

I now have the statistics for 2012/2013 to hand, released without any framing press notice. Here is the updated process behaviour chart.
RailwaySuicides2

Readers should note the following about the chart.

  • Some of the numbers for earlier years have been updated by the statistical authority.
  • I have recalculated natural process limits as there are still no more than 20 annual observations.
  • There is still no signal of improvement or deterioration.

I fear that this is the tough discipline of the chart. It confronts us with current reality and deprives us of the opportunity to find comforting messages. Only a signal on the chart would be evidence of improvement. Statistics are not there to be selectively reported only when they fit our wishes and hopes. Statistics are to be charted, and reported, and discussed, and used as a basis for managing any operation; year in, year out.

Remember that in leading any operation the manager is confined to the retreating picture in the rearview mirror. Without the process behaviour chart, the manager is deprived even of that rear view.

It is a sad picture but improvement only comes from confronting current failure and finding new ways to intervene and redesign. Nobody will benefit from an ultimately vain quest for comforting messages.