A parametric or state of the art analysis is the comparison of existing similar ships to develop preliminary specifications for a vessel during the initial stages of design. As it employs aggregate data, it is not appropriate for designs with few or no equivalents; however, it provides reasonable initial values for ship parameters, and can be continually checked back upon when considering design compromises later in the design process.
As large an aggregate data pool of ship information as possible is compiled for vessels undertaking similar tasks. In particular, carrying capacity (such as tankage or cargo hold sizes, TEU capacity, or number of passengers) is used to limit the range of included ships such that the collection is relevant to the mission profile of the ship under design; though some range in sizes is desirable for the production of design trends. Data should be as recent as possible. Vessels of aged design, or whose performance has been sub-par, should not be included in the datasets. For established yards or design firms, this includes tested data of as-built vessels.
The designer will attempt to locate relationships between design parameters by creating trend curves, seeking (ideally) linear relationships. Design lanes are formed by considering values within 5-10% of the line or curve of best fit. If data is incomplete, a number of estimation formulae may be used to fill-in weak particulars; for example, displacement is often difficult to locate from public resources; so we may employ the Admiralty coefficient to provide estimations of displacement from speed and horsepower (these formulae are generally derived themselves from parametric analysis). Ideal relationships are formed by highly correlated non-dimensional coefficients, such as based off form coefficients or cubic number. This allows for rapid consideration or goal seeking of parameters.
Starting with cargo capacity, values for length, breadth, draught, depth, freeboard, displacement, powering, range, tank capacities, etc. are determined by successive interpolation of trend curves. Ideally, all values should fall within the design lanes on all or most trend curves. Some values may already be known from design constraints, and can be used to help determine starting points for graph evaluation as well. The relative sensitivity of variables should also be noted (i.e., how sensitive is powering to changes in breadth?). With a full understanding on the relationships, the designer will be able to vary particulars and compare potential concepts in an intelligent manner in the concept design phase of the design spiral.
The state of the art analysis also should include an overview of operating capabilities and arrangements of working areas in existing fleets. This is focused on best practices and seafarer familiarity. These vessels also represent competitors for which a value differentiation must be achieved for a successful design. Identification of odd features of a vessel can also be used to inform relationships or exclude particular vessels from parametric datasets in an intelligent manner.
A state of the art report, if created, should include the basis of the investigation (a summary of the mission profile), a brief history of the vessel type culminating with distinguishing features of current designs as well as future trends. This includes unique features, compromises, or equipment load-outs. Indication should be given of the scope and methods employed to gather information (what, and where was, data included, excluded?) though actual source data is often not provided.
Successive design and evaluation will continuously replace derived values with theoretical and actual values from the design vessel. This is particular true for hull form, powering, weight estimates, and capacities. The dataset will likely retain value, however, as general arrangements and the like collected during the research phase can provide insight in the later stages of design. It may also be referred back to at any stage in the design.
It is beneficial to return to the relationship graphs periodically during the design process. If a ships particulars, after further calculation, lie outside of the design lanes (or is a particular outlier), it may signal an problem with the design or highlight a component of the design which could use particular attention. Some departures may be expected, and thus would be acceptable (i.e., an icebreaking tanker would be expected to have a higher displacement, steel weight, cost, etc. for the same capacity as a tanker without ice consideration).