Merchies - How to View Them

[JWE]

There are lots of different ways of characterizing a cargo ship. The most prevalent is in terms of its Gross Register Tonnage (GRT). Everyone knows (or should know) that this is a bogus value dealing only with a vessel’s insurance and canal toll specifications. It’s very reasonable for evaluating a mass of disjoint vessels, but doesn’t help figure out who can actually carry what.

What a vessel can actually carry is quite different from her GRT numbers. A lot is predicated on her design: designs were made to try to beat the toll requirements, so you have “shelter deck”, “scantling deck”, and “full scantling” designs, all capable of carrying a 60% load of DwT, so as to float at “summer, NA, Plimsol” without violating insurance constraints (the weather, sea state, and temperature parameters of Lloyds values are outside this discussion, but send me a pm, and I’ll send you the data) but having very different net cargo deadweight figures.

Thus, three exterior identical vessels, but with shelter, scantling, and full, sections, may have similar GRT numbers, but the ‘scantling deck’ version would sink under overload conditions that would make a ‘full scantling’ design just wallow.

What the shipping companies, their ship masters, supercargos, port factors, and route administrators did was designate vessels, to anticipated routes, based on their actual weight capacity (denoted as CDwT) or their actual volumetric hold capacity. All depended on whether the anticipated cargo was weight limited (ore, raw oil, cerials), or space limited (toilet paper, indigo, oranges, gutta percha). Some vessels had a reefer capacity, for Argentine beef for example. All this makes the ‘honest to God’ capacities of individual various vessels, difficult to get a handle on.

More at next.

[JWE]

Volumetric capacity is different from NRT. For internet drenched auditors, just look at http://www.apl.com/history/timeline/stat6.htm where you will find the actual cargo hold capacity, in cubic feet, of a number of wartime vessels, along with their displacement, deadweight and gross tonnages, as calculated by APL in accord with 1947 standards. Please note these are way different from their GRT valuations, and have nothing to do with their CDwT figures.

CDwT is a measure of what weight of cargo a ship can carry, before sinking. Much has to do with her design (a full scantling ship will carry more, overloaded, than a shelter deck ship). Hold capacity and CDwT are quantities that ship masters, supercargos, port factors, and route administrators used to determine what might be loaded onto which vessel (and where, because of port loading constraints), at what times.

This is data that is very informative to shipping interests, but not interesting to anyone else, so is not your typical internet poop. Rather, the data comes from old fashioned leg-work; design specs, building yard trials, supercargo diaries, and the like.

[JWE]

Naval architects have a fairly good grasp on what a hull might hold, given the ship specs. Since the hundreds of individual ships had hundreds of individual specs, one might reasonably unify the values by statistical analysis. Thus, a statistically significant collection of 2500 GRT vessels might well have a statistically significant aggregate average value of 1250 tons CDwT.

So given all this mess, it is a relatively simple proposition to develop CDwT figures for a class of ships. Obviously, it doesn’t work for coal fired vessels, or composite fired vessels, or diesel vessels, but these parameters are so well known to the architectural community, that they can be accommodated with a standard constant, known in the industry as K’.

[JWE]

In terms of tonnages, register tonnages are useful in judging the relative capabilities of merchant fleets; that’s one reason they were developed. They aren’t very useful in determining the carrying capacities of a vessel.

Deadweight tonnage, often abbreviated as DwT, is the difference between a vessel’s fully loaded displacement and it’s “light” displacement; problem is there were 7 different measurement standards for light displacement, the most common of which were USMC Light, US Navy Light, and Lloyds Light, all different, so DwT is also a moving target.

Net Cargo Deadweight tonnage, often abbreviated as CDwT, is what a ship could effectively load before sinking. Doesn’t matter if its feathers or lead, or in the hold or on deck; a ton is a ton and displaces that much water. A vessel’s CDwT is on the top of the cargo ledger for every vessel ever made. Every ship’s master had one, as well as every supercargo, port factor and all major mercantile port authorities. If a vessel ever called into NY or LA (or any N.A., S.A., or European port or any other port under Euro administration), complete records of her CDwT, cargos, hold allocations, waterline trim (in and out), and actual hold capacity in cubic feet, are listed in the port records.

US and Commonwealth maritime law compels shipping companies to register the design specifications of all class-1, 2 and 3 vessels, seeking entry, with the appropriate national mercantile or trade bureaus. You want the design specs on the Sado Maru, well, the MARAD microfiches at the Library of Congress has the detailed NYK build papers (all 544 pages of them). The MARAD fiche area of the LoC has catalog entries for over 350 individual Japanese vessels for the period 1925 – 1939.

It’s even easier than easy to get this stuff because the LoC research librarians will search, retrieve, vet, and preview the material for you; presuming, of course, you can give them adequate search parameters, tell them the stack range and have paid your annual dues.

Given all this, it is a mathematical exercise to derive a weighted statistical average of CDwT for any class or grouping (30 or more) of cargo ships with similar gross tonnages. Statistical weighting depends on the distribution of scantling specs across the sample, and it might be wise to aggregate vessels in terms of power plant design (oil, coal, composit, diesel) and specific plant BHP, and evaluate their mean/sigmas independently.

Bottom line – it’s not hard at all. If you know where to look, what to look for, and what to do with what you get, it’s pretty easy. Most supercargos and port factors had the equivalent of a 10th grade education. They just happened to be “in the zone” for that little piece of commerce.

[el cid again]

All of which is nice and good - but ignores the problem that many of our vessels are not properly identified - that many of them have no records extant to look for if they could be properly identified - and that for the ships that we might be able to identify, the sheer volume of work would require man-years. It took AKWarrior a thousand hours or more to look up dates of entry to PTO for our LSTs - a small set compared to the merchant ships - and this from a single source. Single sources are indeed best - because it insures common standards of data - and because once you have it - you have it. If this was really "easy" to do - it would already have been done (by AKWarrior for one, by Matrix itself, and by myself). I still say - if it really is easy - then just do it for us and post the data with sources. Or prepare a spreadsheet from which we can copy it electronically. Bluff is still called. The really interesting one I am waiting for is how you are going to get data for a ship that never entered a US port, was not registered with Lloyds, whose builder and owner no longer exist (or no longer have extent records)? We have hundreds of these - probably more than USN LSTs.

[Buck Beach]

ORIGINAL: el cid again

All of which is nice and good - but ignores the problem that many of our vessels are not properly identified - that many of them have no records extant to look for if they could be properly identified - and that for the ships that we might be able to identify, the sheer volume of work would require man-years. It took AKWarrior a thousand hours or more to look up dates of entry to PTO for our LSTs - a small set compared to the merchant ships - and this from a single source. Single sources are indeed best - because it insures common standards of data - and because once you have it - you have it. If this was really "easy" to do - it would already have been done (by AKWarrior for one, by Matrix itself, and by myself). I still say - if it really is easy - then just do it for us and post the data with sources. Or prepare a spreadsheet from which we can copy it electronically. Bluff is still called. The really interesting one I am waiting for is how you are going to get data for a ship that never entered a US port, was not registered with Lloyds, whose builder and owner no longer exist (or no longer have extent records)? We have hundreds of these - probably more than USN LSTs.

Right or wrong Sid you are being dogmatic and talking to a deft audience. IMO you are tweaking him. Leave him alone in his crusade and benefit from his experience and research. He does not want to hear anything you have to say or bring to the party. Short sighted by him but his right.

[Buck Beach]

Experts should feed off of each other and not butt heads as then they learn nothing or discover errors in their opinions.

[wworld7]

Perhaps it would have been better to PM the people that upset you, rather than the tack that you took. IMHO it would have been a more positive way to express your thoughts.

[Buck Beach]

ORIGINAL: flipperwasirish

Perhaps it would have been better to PM the people that upset you, rather than the tack that you took. IMHO it would have been a more positive way to express your thoughts.

Actually a public exposer may be more effective. I am sure others have observed the same issuse, but , thank you for your suggestion.

[JWE]

ORIGINAL: el cid again

All of which is nice and good - but ignores the problem that many of our vessels are not properly identified - that many of them have no records extant to look for if they could be properly identified - and that for the ships that we might be able to identify, the sheer volume of work would require man-years. It took AKWarrior a thousand hours or more to look up dates of entry to PTO for our LSTs - a small set compared to the merchant ships - and this from a single source. Single sources are indeed best - because it insures common standards of data - and because once you have it - you have it. If this was really "easy" to do - it would already have been done (by AKWarrior for one, by Matrix itself, and by myself). I still say - if it really is easy - then just do it for us and post the data with sources. Or prepare a spreadsheet from which we can copy it electronically. Bluff is still called. The really interesting one I am waiting for is how you are going to get data for a ship that never entered a US port, was not registered with Lloyds, whose builder and owner no longer exist (or no longer have extent records)? We have hundreds of these - probably more than USN LSTs.

I have no idea what you are talking about. The premise of the thread goes like this:

1) a statistically relevant sample size of golf balls will give you a statistically significant numbers of dimples per unit surface area;
2) a statistically relevant sample size of tennis balls will give you a statistically significant value for a tennis ball’s surface area;
3) math will tell you how many golf ball dimples will therefore fit on a tennis ball;

What this means is.

a) aggregate merchant fleet data is conventionally given in terms of gross register tonnage (oranges); for purposes of fleet comparison, it is sensible to compare oranges to oranges;
b) gross register tonnage does not express a vessel’s actual cargo capacity (grapefruit);
c) a statistically significant sample of vessels grt and cdwt values, will yield a statistically valid correlation between GRT and CDwT;
d) thus, one may calculate how many grapefruit will fit into a box, once one knows how many oranges it will hold.
e) thus, it is valid to discuss global parameters like merchant fleet composition in terms of GRT;
f) thus, it is equally valid to correlate an average GRT figure to an average CDwT figure.

So, one is entitled to discuss Japanese wartime shipbuilding in terms of aggregate and average gross tonnages, and also maintain a clear understanding of what that meant in terms of vessel size and cargo capacity.

[el cid again]

This is more or less correct in many of its isolated details. And by listing so many points, it shows the "solutions" are not "easily obtainable" - as was alleged (for reasons I do not understand).

There is, however, at the core, a fundamental flaw: the obsession with weight. Here I will confess that is my own preference. A graduate of a damage control school with a gold certificate (that means perfect written and practical scores), I tend to like to think in terms of weight, because when total ship weight = displacement, it sinks. But in the real world of military shipping, cargo is not generally weight limited. A troop seems to require a ton in WITP code, and that is reasonable (although unfair to Japan who put 4 men in that space on an AK in a "Tsuji box"), but a troop does not weigh a ton - if you see what I mean.

Part of the problem is that we don't have enough kinds of ships. 2/3 of all cargo moved to wartime Japan during a year the economy was semi-functional (that is, 1942) was coal. A lot of the rest of the cargo was iron ore. These cargos are indeed very heavy - and weight limiting would be a proper approach for bulk carriers that carry them. But a great deal of the rest of economic shipping, and almost all military shipping, is space limited. [The Allies are not running as much of their economy on the map, and do not move nearly as great a fraction of coal or iron ore. And I found we cannot actually use all the resources really produced and shipped - our factories consume them at the wrong ratio to fuel - so fully 1/2 the AKs are removed - to get the ratio to fuel required in sync - and from a player point of view 2/3 of the ships are not moving coal at all - players use oil where coal was really the fuel of much industry. I have the impression the original model was not wholly thought through.] And we can put either cargo on any AK, although we cannot put resources on amphibious vessels (never mind you really could carry resources in the hold of my 10,000 ton APA, at least some kinds of resources, which was not IRL very different from the AK she was born as: USS Francis Marion was built as the Prarie Mariner). Curiously enough, the idea that a "ton" is space is more true than that it is weight - the way most of the critical cargo to military operations is done - although I would prefer to think of it in terms of cubic meters instead of 100 cubic feet per ton. [A cubic meter of water weighs a ton; if you fill a cubic meter of cargo with most cargos, you will not sink the ship, because it weighs less than a ton; but you cannot fill the ship with more cubic meters of stuff than there is space to put it in - except of course you can add stuff on deck - but if you do - you need to worry about the sum of the weights]

Another problem is that different kinds of propulsion interact with cargo efficiency and fuel efficiency - and the game system is not ideal here. We cannot fire a ship with coal. Yet coal and mixed fired ships exist. CHS and RHS let you fuel them with fuel - but only half the fuel required by an oil fuel ship. Yet the actual coal takes more space and reduces the cargo for any given tonnage of ship. Diesel ships are also very fuel efficient - but this is better represented - we can use actual fuel weight and actual range for them. But they don't need as much machinery space - and may be more cargo efficient. So we might want to modify cargo ratings to account for ship propulsion types.

A statistical approach is exactly what all forms of WITP have used. I thought you were saying it was incorrect, and we could "easily" use actual data. But now I see you have come to saying we need to use statistical methods after all - which means that the earlier apprent criticism should be ignored. But how do we decide what the data should be? AKWarrior worked up some data from official US Maritime statistics, and we used that as a basis for rating all ships. Tankers seem easier - their fuel or oil cargo is a bit more predictable - and even for Japan we often know exactly what the cargo and ships fuel data is precisely. If the cargo of a tanker is less dense than water (it always is), it is always close to the same value, so we can use a statistical approach very productively. I think you will find that is already done, and you won't disagree a whit with any tanker rating for cargo or ships fuel requirement: usually it is exactly correct for some ship - and that ship is used for all similar ships. The question is, can we do this for AKs? And, if we can, what about APs? They are quite tricky - particularly in the context of our code: an AP cannot carry resources, but it can carry supplies and troops; we don't get to decide what the loading ratios to cargo definition is however. But in principle, it ought to be possible to work out a reasonable rating. We may have done both these things already - to the extent we started with good data. OTH in many cases we only know the name and grt rating of a ship - I consider it lucky if we know the engine type and speed as well - and lacking details - how do we decide which "class" of ship we have good data for to put it in? I found many cases where ships were wrongly classified - either as the wrong type of ship - or as the wrong size of ship. But I still have to guess which specific class to put it in most of the time - unless I happen to know it is Gokoku Maru class or something specific.

[keeferon01]

Hi JWE , do you work in shipping, I am quite enjoying reading your
posts om merchants, the reason being my dad was in shipping for over 45 years
he has retired now, he was a stevedore , he would fly to various ports around the
world and train them on different vessels on how to load them , he went to Hong Kong alot,
good stuff.

ronnie

[Dili]

Many thanks, any info will help me in WITM 1940. I am not sure if this was posted but Digitised Lloyds Register from 1930 to 1945 is here http://www.plimsollshipdata.org/  (it ofcourse excludes axis ships build in war years)

also

World Ship List w/ over >200000 ship recorded. Includes war production.

http://www.miramarshipindex.org.nz/

[JWE]

ORIGINAL: Ron James

Hi JWE , do you work in shipping, I am quite enjoying reading your
posts om merchants, the reason being my dad was in shipping for over 45 years
he has retired now, he was a stevedore , he would fly to various ports around the
world and train them on different vessels on how to load them , he went to Hong Kong alot,
good stuff.

ronnie

Never worked in the industry. Grew up in Tampa, FL with Chuck & Zoe Lykes and Geo. Blaine Howell. Been a Lykes Lines barnacle ever since. Took some Nav.Art. in college, and helped design hot-rod sailboats. Was a Govt. pencilneck and did some amphib analysis, mostly Warsaw Pact and ChiCom. Mostly, I talked with folks like your dad.

A little of this, a little of that, but I do love 'em.

BTW Dili, great sites !!!! Very well done !!!! Thank you.

[Dili]

Glad you liked [:)]