Sunday, August 07, 2005

On detecting bad design...

The process employed in an oil refinery is, essentially, a very simple one. Crude oil is placed into large vessels which heat the crude and cook the various products out of it (be it jet fuel, diesel, etc.). Of course, the quality of the crude, plays a major role in determining just how it is cooked.

Connecting the many vessels and tanks in an oil refinery is a labyrinth of piping. This piping is not, as you might imagine, of the type you would find at your local Home Depot. It is specialized piping made of specialized materials, depending on the particular application in the refinery. While many of the components of a refinery are similar, their configuration is typically unique. As such, the piping required is designed specifically for the refinery. These pipe pieces are referred to as spools because they have flanges at both ends and, consequently, resemble a common spool.

A cross-section of a typical pipe spool.

Certainly, the complexities found in a refinery will result in pipe spools of various lengths and sizes. It is also not unusual to find pipe spools with bends in them, such as shown below.

Now, due to the nature of unique design, as found in virtually all refinery construction, one would expect that a methodology that employed a common template to be a better design, as compared with a methodology that designed many, if not all, of the spools individually. For example, if one were confronted with several design problems in which the solution turned out to be a common pipe spool of the size and configuration shown above, wouldn't it make sense to keep that configuration? After all, it would save money in terms of design and fabrication time.

Conversely, if one found out that, instead of employing a common template, a unique configuration was used for certain spools, would that be cause to infer bad design?

No. It would not. And the reason why is that one may not have complete access to the design parameters at hand.

Recently there was an explosion at a BP plant in Texas City, Texas. In BP: Wrong type of pipe caused plant explosion, we read,

The installation of the wrong type of pipe spool is believed to have caused last week’s explosion and fire at the same BP plant where 15 people were killed in March, company officials said Monday.

An 8-inch section of heavy steel pipe, located between a compressor and heat exchanger on the Resid Hydrotreater Unit, failed on Thursday. No one was injured during the resulting blast and blaze. The section failed because workers installed the wrong type of steel pipe spool on the outlet of the heat exchanger, BP spokesman Ronnie Chappell said. The elbow-shaped pipe spool, believed to have been mistakenly installed when the unit underwent routine maintenance in February, is not designed to withstand higher temperatures.

Carbon steel pipe is used for lower temperature operations, while stainless steel pipe is used for higher temperature operations. Installing a carbon steel pipe spool in place of a stainless steel pipe spool will result in pipe failure.

Installation procedures should have prevented the installation of the wrong pipe spool. But human error is a tricky thing. So tricky, in fact, that one would expect a better design to be one that takes into account the possibility that a pipe spool of the wrong material type might be inadvertently installed due to its identical configuration with the correct spool.

The lesson here is not that it's difficult to detect design. No, that's the easy part. What's difficult is to conclusively determine that the design detected is truly a bad design.

Keep that in mind whenever you hear evolutionists make claims that certain biological systems or organisms exhibit inherently bad design.

1 comment:

Anonymous said...

Great analogy!