Tuesday, 8 May 2012

The Prisoners' Dilemma

One of my colleagues has recommended this video showing two contestants faced with a game theory decision similar to the prisoners' dilemma.

Monday, 7 May 2012

Do O.R. people make business entrepreneurs?

Although I live in the U.K., I subscribe to the magazine OR/MS Today, which comes from INFORMS, the US equivalent of the UK OR Society.  Besides having news and interesting articles, there are regular columnists, reflecting on our discipline.  In April's issue, the column (Oracle) by Doug Samuelson is headed "The Runner's Parable".  (It is only accessible to subscribers.)

In the parable, one of the characters says: "Generating new ideas and avoiding mistakes are two completely different management styles, and you can't do both at the same time."  When I read that, it made me think about the place of the OR scientist in a management structure.  Which style corresponds to most OR work?  I suspect that the answer is in the latter style - avoiding mistakes.  That is not to say that OR work does not sometimes deal with "new ideas", but generally OR is concerned with improving and investigating the current work of management, and helping to avoid mistakes.

I wondered whether Doug S had taken this sentence from some other source, but a search online did not find it.  Maybe it appears in print somewhere?  Whether or not it is part of some established management philosophy, it is an astute observation.  I think of the friends in management positions, and recognise that very few actually combine the skills to innovate and avoid problems.  Perhaps some of the world's economic ills at present are the result of too few senior managers with the style of "avoiding mistakes"?

On a number of occasions, I have been part of a discussion about the career paths of those who start as OR scientists.  (The proportion of people with OR qualifications who spend their whole career in the discipline is quite small.)  Many go on to management roles, but few are committed to innovation.  I have asked whether there are OR people who could be classed as entrepreneurs.  And generally, the answer is that entrepreneurs have a different outlook on business from that taken by someone with an OR philosophy.

I may be wrong, but I suspect that OR is at the opposite end of some management spectrum from entrepreneurs and innovative managers.

Thursday, 3 May 2012

Supply chain problems

A long time ago I wrote a review of a book called "Inventory Management in Lebanon" for the Journal of the O.R. Society.  I doubt if anybody bought the book as a result of the review, but I felt that it deserved a review in the world-wide O.R. literature.  It had been written by an academic at the American University of Beirut, and published by that university.  I read it while Tina and I lived in the neighbouring country of Jordan.  My students there, all on a part-time MBA course, were aware of the problems of being at the end of a complex supply chain for items that had to be imported from North America or Europe.  The book discussed conventional models of inventory control, and explained where they fell down in Lebanon (and as I saw, in Jordan as well).  Lead times were erratic, items deteriorated while in stock, there were government controls on overseas purchases and import restrictions.  This was the reality which did not actually match the theory in most O.R. textbooks.

Last week, Tina and I encountered a similar problem, rather closer to home.  We were on a short holiday on the island of Jersey in the British Channel Islands.  Each day we ate our lunch out.  One day was extremely wild and wet, so that ferries to the islands were cancelled and so were some flights.  The following day we went into the local branch of a national chain store to try and use a coupon for their sandwiches, to find there were none.  That chain store sourced its sandwiches from the mainland.  Their supply had failed because of the weather.  Echoes of the erratic lead times of Lebanon.  We bought sandwiches made on the island (presumably the bread was sourced in the island) which were much more pleasant (and more expensive). 


Mont Orgueil Castle on Jersey

So, Tina and I wondered about the decisions made by that chain store about its supply to Jersey.  We concluded that the supply chain must be sufficiently reliable for the company to continue its practice of supplying the shop from the mainland, otherwise they would have chosen to source these items from a local source.  And we wondered if there were any items where they held larger stocks in their Jersey branch than they would for a similar sized branch on the mainland.  (No, these speculations didn't spoil our holiday!)  And, by the way, I would be happy to return to Jersey if any company would pay for Tina and I to go there to research supply chain problems there!

Infrastructure decisions

A long time ago I worked with a local authority on a planning problem.  It was my first foray into problems of long range planning of that nature, and I had a great deal to learn.  I came away with several lessons for my teaching, and an increased awareness of how the infrastructure of a town is interconnected.

That episode was concerned with the development of a primary school to serve a large new housing development.  The simple question was how large to make the school.  But, it is relatively easy to decide on the ultimate size of a school, given a catchment for its pupils, but it makes sense to have a smaller school initially, and expand later.  At one stage we had a three-by-three matrix, with three sizes of school, and three scenarios for the demand in the first phase of the housing development.  Each of the nine cells had practical and political consequences.  It was a useful tool for thinking with.  To make the planning more complex, the estimated time for planning and constructing a school was about four to five years.  So, one might estimate the demand from the number of children on the estate ... except that families with pre-school children are (or were then) extremely mobile between houses.  So many of the children in the estate in year 0 will not be there in year 4 or 5. 

My home city of Exeter, and the county of Devon, are facing a similar problem, but one which is potentially more serious.  It is concerned with the provision of hospital facilities.  We have one large general hospital in the city.  (As an aside, the estate agent's publicity for the house where we live stated that it was "convenient for schools and hospitals".  The second comment was intended to encourage staff from the hospitals, and was not aimed at hypochondriacs or the long-term ill.  As it happens, we have one hospital consultant living close to us, and two more former consultants as well.)  It was built in the 1990s to serve the population of the city and the country around.  The problem now is that the population in and close to Exeter is set to rise with new houses, new estates, and a new town (Cranbrook), a few miles outside Exeter in the hospital catchment.  And as the population grows, so does the need for hospital facilities.  Already there are cases of "bed-blocking" when patients cannot be admitted because the wards are full.

So, what can be done?  There is not much scope for expanding the hospital on its present site.  There is scope for changing the regime of hospital treatment, and this is already happening.  Some patients can be discharged to smaller hospitals which specialise in post-operative care, but the expectation is that there will be increasing problems of capacity in the existing hospital.  And this is a situation which is repeated in several parts of the county, with four large general hospitals (Exeter, Plymouth, Torquay and North Devon).  There is a limit to how much the demand can be handled by changes to the regime.  There's no fifth large conurbation where a hospital could be sensibly placed. 

As an O.R. person, it is interesting to see how hospital management has considered decisions relating to the wider system, and not simply the capacity of the building.  The medical services work to reduce the need for patients to spend time in hospital (recalling the worrying statistics about how many patients catch diseases while in a ward overnight), carefully schedule the patients who can be scheduled, plan operations and the recovery afterwards.  O.R. people should always be encouraged to think of the whole system - it is not a problem of capacity alone.  The discussion that I have seen has not broached the possibility of building a new hospital yet.  I wonder how long it will be before this happens?