• [Web] Service Versioning Guidance:
  
  Keywords: SOA
• Making Your Application a Windows Vista Application:
  The Top Ten Things to Do
  Keywords: Windows/Vista
• JavaScript tutorial:
  Simon Willison’s tutorial from ETEch
  Keywords: Javascript
• Ajax Templating:
  Seperating layout and logic with Ajax
  Keywords: Web_Architecture
• The challenges of complex IT projects:
  BCS position paper
  Keywords: Project_Management
• Strategic Knowledge Management: using the Theory of Constraints for Better Knowledge Management:
  Reinventing knowledge management through TOC
  Keywords: KM, TOC

Clarke quotes the introduction which sets out the way the methodology was adapted, within the body of the document there is more on how they assessed capacity at the various constraints along the route.

Another quote relates to how the first list of probable constraints was found:

The starting point of the analysis was the selection of the potential constraints to be considered. A draft list of locations to be studied was produced using Network Rail’s detailed working knowledge of the East Coast Main Line. One approach has been to discuss potential issues with the Network Rail Timing Specialists who have a timetabling responsibility for the area of interest. Commonly they have been asked ‘which location / locations are likely to prove problematic?’. One of the tenets of the ThOC is that the over-riding constraints will be widely recognized. The locations selected for analysis have been discussed with each of their operators and they have each indicated their agreement. It is reassuring in this respect that the locations studied to date feature prominently in the responses of consultees to the applications.

Although the document is 85 pages long, I think it is worth at least a skim as an example of application of the approach to a new field.

What seems especially valuable to me in this example is the evidence that even in the complex multi-company and government environment of the UK railways, the use of a logical approach such as the Theory Of Constraints was capable of gaining support from the parties involved.

It is this aspect which, I think, points to the wider usefulness of the example as a learning tool which may point to new areas where application of constraints thinking could be useful. For example I can see possible application in the media production industry where application of digital networked techniques to complex supply chains will inevitably lead to contention over common pieces of infrastructure.

Summary notes of the problem are here.

The challenge is to make the system profitable, with a strong steer to focus on increasing revenue. This post contains my first thoughts about a solution.

My first thought was to look at the constraints in the system – how could the management increase Throughput without increasing Operating Expense or Investment?

It would appear that there are different constraints at different times of the year.

In the peak summer season the trams run near capacity at all times, suggesting that there is excess demand, and the constraint is in the contribution received for each passenger carried. An easy strategy to try here would be to increase the fare price and thus the contribution per passenger carried. The case asserts that for the affluent tourists the current fare is insignificant, so the market should bear this.

In the early and late weeks there is excess capacity on the trams that run, suggesting that the market is the constraint. The passengers are mostly locals, and are price-sensitive. A 20% price rise has created a drop of 40% in passenger numbers in the early weeks of the season. If this is reversible then reducing prices in the off-peak part of the season should be offset by increased passenger numbers.

So strategy 1 is seasonally-adjusted pricing, with a reduction in the off-peak weeks and an increase during peak periods. There is a policy constraint that requires the average fare across the season to remain at $2.

Assuming that the price-sensitive drop in passenger numbers is reversible, then initial analysis suggests that reducing the price to $1.66 (i.e. a reduction to prices from 20 years ago) in weeks 1-13 and 25-32, combined with a price rise to $2.50 in weeks 17,18, 22-24 and to $2.60 in weeks 19-21 (the increases calculated to meet the average price constraint) will move the company to profitability, even accounting for the loss of state subsidy, wage increases and loan repayments.

To sense-check this would need some detailed figures on capacity which are not available in the study.

In the comments to Ming’s post there are a range of views expressed but two themes emerge:

* using the tools available today to _filter_ incoming information and tasks, allowing you to concentrate on the important things
* there is indeed a very sharp limit to the power of conscious processing to handle multiple tasks (Miller’s [bliki]SevenPlusOrMinusTwo[/bliki]) but the unconscious mind is capable of many many simultaneous activities.

From my own subjective experience I would suggest that one reason why having too many things to do “simultaneously” hits productivity is because it ignores the way the mind transfers things into unconscious processing.

The trick seems to be to concentrate on one thing sufficiently long that you build up a whole set of pathways relating to it, then “put it down” and move on to something else – the unconscious will still be working away. Do this and you will be surprised how often the answer “just appears” a few hours or days later.

Time-slicing too finely in the conscious domain seems to have the effect that no topic creates enough energy to engage the unconscious learning circuits, so I’m left relying on the distractable power of the conscious alone.

It would be interesting to explore the neuroscience of this a bit further…

From a [bliki]TheoryOfConstraints[/bliki] perspective it would appear that conscious attention is the constraint, so useful questions to consider might be:

* How do I get the most out of my conscious processing power?
* What else do I have to change to allow my conscious attention to work at its best?
* How can I find other ways of processing information (e.g. exploiting my unconscious mind)?

The entire CRT(Current Reality Tree) so far can be seen in this PDF

Having discovered the Twiki Draw plugin, I think a wiki with a drawing tool could well be the perfect tool for developing this sort of exchange…

In other words by extracting the key concepts from the book into a CRT it should be possible to graphically display and test the book’s argument at the same time as comprehending it.

In the first chapter Argyris gives some strong clues about the sort of Undesirable Effects (UDEs) we might see in the real world…

Argyris gives examples of seven symptoms commonly seen in organisations:

1. Actions intended to increase understanding and trust often produce misunderstanding and mistrust
2. Blaming others or the system for poor decisions
3. Organisational inertia: The tried and proven ways of doing things dominate organisational life
4. Upward communications for difficult issues are often lacking
5. Budget ganmes are necessary evils
6. People do not behave reasonably, even when it is in their best interest
7. The management team is often a myth

He then suggests that for rational, well-meaning human beings to consistently create these sort of problems there must be something wrong with their thinking processes, especially when dealing with business issues that are embarrassing or threatening – they must be using what he calls “Defensive Reasoning” – the three symptoms of which are:

• Individuals hold premises the validity of which is questionable yet they think it is not
• Individuals make inferences that do not necessarily follow from the premises yet they think they do
• Individuals reach conclusions that they believe they have tested carefully yet they have not (because the way they have been framed makes them untestable)

Using the terminology of the TOC thinking processes I’m going to take these as the initial UDEs
Argyris states that the causes of this defensive reasoning are four-fold:

1. The human programs held by the people concerned about dealing with embarrassment or threat
2. The fact that they use those programs skillfully
3. The organisational defence routines that result
4. The organisational “fancy footwork” used to protect the defensive routines

We can use those UDEs and causes to start making a skeleton CRT.

At the moment the the logical jumps between the entities seem too large to start plotting cause-effect arrows; as I work through the following chapters of the book I’ll develop the tree in line with Argyris’s argument.

Earlier articles:

* Invisible dogma and learning organisations
* Learning Organisations and TOC pt 2

Confirmation that my intuition may have taken me down a fruitful path comes from Chapter 1 “Puzzles”:

“The players in these studies also take for granted policies and practices that are contrary to their managerial stewardship. They bypass root causes. They equate being realistic with being simplistic. They make all these actions undiscussable. They thus wind up creating a world in which the bad is tied up with the good so that producing the latter guarantees the former. Finally all of this is done with the best of intentions”

Which also sounds like the sort of situation a CRT was designed to explore!