Understanding Variation by Donald J Wheeler, Chapter 2 Notes

Notes from Chapter 2 of Understanding Variation: The Key to Managing Chaos, by Donald J Wheeler

The chapter is titled:"Knowledge is Orderly and Cumulative". Here Wheeler discusses three approaches to interpreting data, and the benefit of the control chart.

Two people may have access to the same set of data, but they can arrive at different conclusions. The reason is because they may have different interpretation processes.

Not all interpretation processes are valid. Wheeler describes two problematic approaches, and then discusses Shewhart's approach.

The Specification Approach

The first approach is 'comparison to specifications', or the Specification Approach. In this approach, data is interpreted in comparison to specifications, and its 'goodness' or 'badness' is determined based on how well it conforms to specifications.

For example, management may demand that the factory produce 1,000 units per month (the specification). If the factory produces 800 units in one month, that figure is interpreted as 'bad'.

Comparison by specification leads to binary results: either data conforms or it does not. it also leads to sudden changes in goodness or badness. A process could be in good graces one month, and in bad graces in another.

But the biggest problem with this approach is because specifications are the 'voice of the customer'. This is what the 'customer' wants.  The process has nothing to do with it. The voice of the process is what the process is able to achieve. If the voice of the customer and the voice of the process is not aligned, the result is people distorting the system or distorting the data in order to meet the specification on paper.

Nevertheless, the voice of the customer is important because it tells you when you're in trouble. 

The Average Value Approach

The Average Value approach compares the goodness of data against the average value of the data. "Why is sales 10% below last year's average?". This approach to interpreting data leads to pathologies similar to the Specification Approach -- people tend to distort the system or distort the data so they don't have to explain why the data is too far away from the average.

At least the Average Value approach uses the voice of the process (after all, the average value is produced by the process, and not some arbitrary number).

Shewhart's Approach to Interpreting Data

Walter Shewhart was the first to define the voice of the process. He called this a "control chart."

A control charts has time in the X-axis, has a central line, has two control limits, one on either side of the central line, both the same distance from the central line. The distance of the control limits is determined from the data, from the voice of the process. The values plotted can be raw data, or some value calculated from the raw data.

The control chart helps interpret the data by characterising the behaviour of the data and allowing us to predict the future behaviour of the data.

The control chart shows that there really is no distinction between good outcomes and bad outcomes; they both came from the same process.

A process that is in control is predictable. In other words, we can predict its behaviour in the future and make plans based on that prediction.

The Second Principle for Understanding Data

Shewhart's second principle for understanding data is:

While every data set contains noise, some data sets may contain signals. Therefore, before you can detect a signal within any given data set, you must first filter out the noise.

(His first principle is introduced in Chapter 1. It's: No data have meaning apart from their context.")

The two mistakes that could be made in analysing data is treating random variation as a meaningful departure from the past, and the second mistake is not recognising when data is a sign of change.

Attempting to avoid the first mistake by not reacting to variation can cause you to miss true signals of change, and thus make the second mistake. Attempting to avoid the second mistake by reacting to every signal causes you to make the first mistake

The key benefit of a control chart is it identifies for you data that is noise, and data that is signal.

Juran on Quality by Design (Chapter 2)

Reading Chapter Two of Juran on Quality by Design (paid link)

Chapter 2 "Establish Quality Goals"

NOTES

A goal is an "aimed-at" target. A quality goal is a quality target you are aiming for. The goal would typically include "a number" and a "time table"

Some consider that setting quality goals is part of quality planning, whereas others have the view that quality goals are set outside of quality planning -- the purpose of planning in this case is establishing how to achieve the quality goals.

A vision has no connection to reality until it is translated into quality goals.  Planning is impossible until goals are established.

Tactical quality goals refer to goals designed to meet human (customer) needs. These include product features, process control features, etc.  They are set by middle management in order to meet the strategic goals.

Strategic quality goals are set at the highest level of the organisation. They are often driven by the company's strategic vision and policies. Examples of strategic goals are: "Improve product and service quality ten times (Motorola)", and "Reduce billing errors by 90% (Florida Power and Light Company)". 

Upper management should not become involved with individual tactical quality goals as there are too many, but they need to be involved with them at the 'collective' level.

Quality goals do not only come from customer needs. They can be driven by a company's concept which may need to be marketed to customers (an example is Walkman, which no customer 'needed' or asked for). Goals can originate from regulations, or from internal human desires to be meticulous.

Some of the most important bases of quality goals:

• Technology - in the form of specifications and procedures
• Market - quality goals that affect saleability should be based on the market
• Benchmarking - quality goal made with reference to what other companies are achieving
• Historical - performance in the past is used as the basis for performance in the future. Sometimes the goal is to improve on past performance. Sometimes the goal is ensure stability with past performance.  This basis can be a double-edged sword.

Quality goals are a hierarchy. Primary goals are at the top. These include 'personal health' or, for a car, 'effective transportation'.  Secondary goals are essential in achieving the primary goal, tertiary goals help achieve the secondary goals, and so on until each sub-goal is achievable in technological terms.

Planning is an essential part of achieving generating and achieving quality goals. Planning helps identify what sub-goals are needed (by asking: how can we achieve the primary goals?) and how each sub-goal can be achieved.

Deployment of Quality Goals

'Deployment' means allocating the goal to someone who will deliver the goal.

Strategic goals are subdivided in smaller and smaller sub-goals which are then allocated to someone (eg a middle manager). In Japan, that someone determines what resources they need to accomplish the sub-goal. This arrangement enable participative two-way planning.

Strategic goals are too big and systemic to assign to one person. However, it is imperative that every goal be 'deployed' (ie, allocated). One solution is 'teams'. A team is formed and assigned to deliver the goal. The example mentioned in Chapter 2 is 'Team Taurus', tasked to deliver the strategic goal of making Taurus quality best in class.

Provision of Resources

Goals need resources to achieve them. A key blocker to achieving strategic quality goals is the non-provision of resources to achieve them. The practice of developing strategic quality goals seeks to reuse the existing practices in business. Businesses have processes to allocate resources to define, administer, and delivery business goals.  Strategic quality should be approached the same way.

Corporate Interference

Expect resistance from various areas of an organisation when developing strategic goals. The key cause of resistance is the reduction in autonomy of the various divisions and functions. Strategic goals are directions from above, plans to achieve them by the divisions are approved from above, and people from above will be monitoring the division's achievement of those goals.  All three aspects are reductions in the autonomy of the division, and can cause friction, even in the most harmonious relationships.

Idenitfying the Customer

It's curious that Juran says the "next step" after establishing the quality goals, is identifying the customers.  It seems counterintuitive. Wouldn't you need to know your customers first before meaningful quality goals can be established?

Ride Your Bicycle Forward

Organisations with a merit and ranking system are like bicycles being pedaled backward.

 

In the 1950s, W. Edwards Deming, then a relatively unknown statistician, conducted seminars among Japanese companies as part of the American effort to rebuild that country’s post-war economy.  

In these seminars, Deming reportedly used the process diagram shown below, to illustrate the systemic nature of a business’s processes. 

 

The diagram showed how the various parties: suppliers, production, ‘quality control’ (inspection), consumers, research, and others, fed into each other and back in one grand system of production. 

 

Despite the diagram’s deceptive simplicity, it was conveying a message that is both deep and shallow. It was a shallow message because everyone who knew anything about operations knew what it was showing. Yet the message was also very deep because the diagram confronted everyone by asking why they did not act as if they knew that. 

 

Deming made the point that because these functions depended on each other in a systemic way, they must be managed together, as a single system. To manage these functions separately, as if they were not dependent on each other is a path to institutionalised dysfunction.

 

In our modern world of systems thinking, processes management, and post the chaotic period of the 1990s ‘re-engineering’, the reminder Deming wanted to deliver can feel anachronistic.  Not only does the diagram look old, it feels old.  It would not be surprising if a modern audience today reacted to this diagram with a: 'Duh'. 

 

Duh indeed, because by this time we all should all know that. And yet it seems we  don’t.

 

So what is the big deal about Deming’s diagram?  While no one seriously questions its truth, many big companies – precisely the ones that really need to internalise this message – are still operating the business in a backward way.

 

We can illustrate the situation by imagining we were given a bicycle, and then shown how a bicycle should run, and yet we proceed to ride it backwards. 

 

In organisations where a merit and ranking system pervades, where an employee’s ‘performance’ is annually ‘assessed’ by their supervisor,  employees become forced to consider their supervisor as their most important customer.  Employees have no option but to direct all their actions and energies to figuring out what numbers the supervisor is tracking and make sure they meet those numbers, to the subjugation of other considerations.  

 

Supervisors themselves need to please their managers, and so treat their managers as their customer, always thinking: “how am I going to be ranked?”  The situation goes on -- the managers need to please their vice presidents; the vice presidents need to please their executives; executives need to please the CEO; the CEO need to please the board.  The board need to please the shareholders, often fund managers. Fund managers need to please their bosses, in a grand system of brown nosing.

Nowhere in all this is the poor paying customer, the source of the company’s income.  No, that's not entirely true.  The customer does sometimes pop up now and then, but only to the extent where they complain thereby represent a risk to the merit and ranking of employees, or where they give praise and enhance the merit and ranking of employees, or where they can be used to further please that most important of customers, the person higher up in the organisation.

This backward operation of the forward system cannot be undone without a transformation -- a deep, difficult, transformation that demands sustained struggle.  Initiating this transformation, and even more important, sustaining such a transformation will require an unbelievable level of courage, focus, and determination.  A constancy of purpose.

Organisations who are able to operate their forward systems in the right direction will eventually benefit. Once they do start moving forward, it will be easier and easier, kept in motion by a reformed cultural inertia,  to operate the system, understand it, speed it up, improve it, and even accelerate it.  They will have realised how far, FAR better it is to ride a bicycle forward.

Schedule driven projects

How many times have you been involved in programs where the most pressing concern of everyone is meeting the schedule? 

Quality (fitness for purpose in this case) is the first to go.  To hell with quality, we need to get this deliverable --- a document, a piece of software, testing, etc. – finished.

Project status reports often show green if the everything is ‘on schedule’.  There is little status reporting on whether things are being done right.

I was involved in a large multi-million dollar program a few of years ago.  Immediately upon joining the program, I saw how badly the system being built will turn out to be.  For example, data input is being designed without consideration of what output is required (how would we know what input is required if we don’t know what output is required?).

I raised and itemised my concerns, but they were brushed off.  The schedule pressure was of paramount importance.

The result was predictable.  It was a very successful program:

• The program came in on schedule (the nth update of the schedule)
• The program came in under budget (the nth adjustment of the budget of course)

Except for the minor fact that:

• The delivered information system was unusable and unused.
• A second remediation program had to be launched two years later to fix the problems of the successful program.

This second program did give long employment to a second set of consultants and employees.