Page 51 - Practical Ship Design

22 Chapter 1

reservations about data obtained from articles in the technical press and similar
sources. This should not be interpreted as an attack on the technical press which the
author believes does a very good job and is generally accurate, but reflects the fact
that with the best will, errors do occur.
The snag with in-house data is that it is unlikely to be very large, except in a few
very exceptional companies and even in these will probably be limited in its
diversity. Naval architects therefore find it necessary to use other sources but when
they do so should try to reduce the effect of any errors that there may be in this data
by using (and comparing) several sources and amassing such a quantity of data that
the effect of errors can be minimised by “fairing”.
Some particularly useful sources for the data items suggested are discussed
below.

1.3.3 Ship dimensions

1.3.3.1 Data from Lloyd’s Register
Lloyd’s Register is a prime source of data on the dimensions of almost all types of
ship but suffers from the disadvantage that it is arranged in alphabetical order of
ship’s name with old and new ships and all types of ship mixed up.
When writing Chapter 3 the author wanted to use data from recent ships and
saved a lot of work by using Lloyd’s “May up-date’’ - the last update before the
publication of a new volume of the register which contains particulars of all the
ships registered for the first time in the year in question.
Unfortunately his wish to have data on the ship’s design deadweight was
frustrated because the deadweight quoted by Lloyds is that at the full draft. In
addition, Lloyds do not quote the load displacement so if this is required it must be
synthesised. In the spreadsheet already discussed in 9 1.2.3, this was done using the
service speed quoted to calculate the Froude Number and this in turn to estimate
the block coefficient using the mean line on the graph (reproduced as Fig. 3.12),
which represented the practice of a large number of naval architects at the time it
was produced and seems to remain close to current practice. The block coefficient
is then used to calculate the displacement, but it should be noted that the method
assumes that the quoted speed is related to the quoted deadweight which may not
be correct and this could introduce some error.
Table 1.1 records information for tankers, bulk carriers, container ships and
refrigerated cargo ships, the only types for which the 1988-89 and 1989-90
volumes seemed to have an adequate statistical sample, and it came as quite a
surprise to note that general cargo ships had almost disappeared as a category
except for small vessels with deadweights of up to about 5000 tonnes. The ships
included in each sample were selected to cover the range of deadweight but in
other respects the sample was arbitrary, although it has to be admitted that there