Snow, cold and outlying data

One of the recurrent problems of production monitoring is to try and determine if a process is out of control. Observations are made at various times, and from these, one is supposed to determine whether or not there are problems.

An article on the website of KNMI (the Dutch Meteorological Institute) asks whether the winter of 2009-2010 was unusual. It is here. The author asks whether the weather was unusual, by looking at various statistics from different parts of the world. What makes the page so interesting is the way that the statistics are considered with and without a model that incorporates global warming. It is a reminder for O.R. workers doing control modelling to make sure that observations are related to the correct underlying model.

O.R. and the Infrastructure (3)

Another thought about the hidden science. In the U.K. (and I guess in many other countries) most traffic lights (whatever you call them) at road junctions are controlled by computer. Detectors are located close to the stop line and also in advance of that line, indicating the presence of vehicles waiting and approaching. (Next time you are cycling or walking past traffic lights, have a look for black tar-covered lines in the tarmacadam, which cover detector wires, or look for miniature radar sets on the lights themselves.) The logic behind the programs that control the lights is developed by O.R. scientists. In the programme about infrastructure "Britain from Above", the presenter visited a traffic control room, where the staff had the power to change lights when their traffic monitoring equipment (including TV cameras) detected congestion. It was left unsaid that most of the time the traffic flow is controlled automatically; the people in the control room had to deal with the exceptions, the unusual. Why can't the computer control be extended to cover these exceptions? Cost and complexity. It would cost too much to build in rules for exceptional cases, which would be complex. It is good O.R. (IMHO) to know when to stop building too complex a model. Besides, traffic control has multiple objectives, and the importance of the different objectives changes with the time of the day and much else.