murphy's law

After an unplanned sequence of clicking on follow up links on web pages, I ended up on a page explaining Murphy's Law. If you want to know, I actually started off at my homepage, ST online, and passed by a Great Nikon Sale page. Coming back to the point. It was pretty surprising to see Murphy's Law (something which we used so carelessly in our daily speech) to actually have a physical meaning in the form of murphy's equation.

"Murphy's Equation"-Joel Pel

* skip this part if you are not interest in the formulation

Here, P_M is the Murphy’s probability that something will go wrong. K_M is Murphy’s constant (equal to one) and F_M is Murphy’s factor, a very small number that can only be calculated on a 386-computer running Windows 3.1. Here we will approximate it by F_M ≈ 0.01. Similar to previous work, I, C, U and F are parameters on a scale of 1-10 describing the situation of interest. I is the importance of the result or outcome of the event, C is the complexity of the system, U is the urgency and F is the frequency. From these 4 parameters, virtually any event can be described. Plugging those numbers into the equation with the appropriate constants will then give the probability of at least one thing going wrong. To get an idea of how this works, lets look at some specific cases.

First, to start with a simple example, lets calculate the probability of the clutch on a 1989 Toyota Tercel ceasing to function 100km from home at night in the middle of a rainstorm. The importance of the clutch working in this situation is obviously high, but no one is dying, so lets estimate an 8. The system is fairly simple compared to other systems in the car, so C=5. It would be nice if the clutch worked soon, so the urgency U would also be around 8. Finally, the clutch only needs to work for one drive home, so the frequency is low, say F=1. Now, putting these parameters in Murphy’s Equation shows that P_M=1. Comparing to experimental data, this number matches exactly, as the clutch did indeed cease to function at this time. Repeating this calculation with I=7 also gives P_M=1, showing the robustness of this equation.

Moving to a more complicated example, lets examine the case where the clutch has failed in the above example, but calculate the probability that the flashlight needed to inspect the clutch doesn’t work, the front half of the hazard lights on the car don’t work and that the trunk where the few tools are kept has 3” of water in it (remember its raining). In this case, Importance is about the same, an 8 or so (its hard to justify higher than this without a life or death situation). Here, C=4, as the flashlight, hazard lights and trunk are all fairly simple systems. Because it is raining and the driver is now stranded, the urgency can be bumped up to a 9, while F=3 (three events need to occur). Putting these values into the equation gives a P_M of 1, which is indeed what did happen, showing that again this equation matches experimental data very well.

* from science creative quarterly

What I had thought of as a pessimist's explication of his world view had metamorphosed into a physical equation.

an equation derived from human's compulsion to measure everything..