Design Phases vs. Design Spiral

Hi,

We often discuss only about the Design Spiral, but I am very interested in knowing how many of you:

• Use only the Design Spiral?
• Instead of using the Design Spiral, use a design process based on Design Phases (for example, Feasibility -> Concept Design -> Advanced Design -> Production Design, etc.).
• Use a combination of both Design Phases and the Design Spiral?
• Do not use any design process or design approach at all?

Also, from your own experience, which type of project (for example, based on complexity, length, displacement, some ratio, ...), one approach or the other, or the combination of both, would be more fitted?

Thank you!

 

In my opinion, after having designed many ships of all types and many sizes, the design spiral is a theoretical concept, very useful to explain what everyone already knows, a very clever concept, which has appeared for centuries in the books, which in practice no one follows. Each designer, in the real world, has their own working method and, what is completely true is that it is a process in which you take three steps forward and one step back.
It's not worth discussing it.

 

In the world of AMV's, "design spiral" has a bad connotation. We refer to a "design spiral" as the often non-converging design process that many ACV, SES, hydrofoil and hybrids thereof are so often subject to....or the victim of. Start at 35m LOA..surface effect ship..nope..customer requirements overload that beyond known constraints on cushion loading etc., so we go to 38m...more "stuff" gets added..so now we're at 40m..but now the cost of the thing exceeds the original funding budget and it dies. That is what we call a "design spiral". ;-) We've seen more of those over the last 35+ years than I could probably recount.

 

After working on many government projects a lot depends on the criticality and the costs. The higher either of those two, the more divergent the solution. Critical systems need more flexibility, expensive systems need more design spiral.

Generally, the sequence is: Identified Need>Brainstorming Studies> Concept Methods vs Design History> Reality Downselect> Feasibility Studies, ROM (Class F) Costs> 1st Downselect> Preliminary Concept Designs> Multiple Low Turn Design Spirals, Class D Costs> 2nd Downselect> Preliminary Design, Class C Costs> High Turn Design Spirals and/or Quarterly Progress Reports> Detail Design, ESWBS Population> Contract Bid...

Of course that is for well managed designs...otherwise I've seen contractors throw stuff at the wall and then see what sticks...

 

I am more in line with these successive steps :
** Feasibility is either to gather the main figures based on know how and statistics from previous similar models if any, or to do a specific study for a prototype inc. novelties with some proof of concept/ demonstration at right (small) scale if need be.
** Concept or say early stage project : at the end the main dimensions, propulsion and payload (for the aimed programme) are fixed (and received agreement from the customer) + the total weight estimated based on statistics.
** Advanced or say Engineering design : mostly all the structural issues are addressed and the total weight is based on documented drawings inc. scantlings + list of equipments.
** Production design, which can also includes « method of » studies, includes the drawings stamped « Good for construction » for the shipyard.
Within each phase, you have iterations which can be called design spirals.
This ideal process can be a lot disturbed by the attitude of the final customer : who can continuously change his objective, the final functions, the list of equipment, the speed, the autonomy, etc … An experimented customer can be a dream for the designer , a beginner customer can be a nightmare.
And the red line all along the process is the weight mastery : a good project can become a poor one at the end due to too much extra weight. Weight centering fore/aft is also to take care.

 

The Design Spiral in "The Principles of Yacht Design" is the first illustration in Chapter 1.
The Design Spiral can produce intuitions and assess feed back loops to assist us in the design process.
"Intuition Pumps are Tools for Thinking" We humans do this kind of thinking all the time but without some sort of structure/Design Phases it is only a "Free Floating Rationale" methodology at best and "like Darwin Theory of Evolution it can be slow to change and can lead to many dead ends."
-Daniel Dennett

I try to apply Kant's Judgements to each spoke of the Design Spiral and add some sub-sets of my own if I think they will reveal some insight into the design or process.
-Quantity - the amount or number of something, especially that can be measured
-Quality - of something can be determined by comparing a set of inherent characteristics with a set of requirements
-Modus operandi - an individual's habits of working, particularly in the context of business or work
-Relationship - Are they on the same page as you and your goals, things or people that are willing to help you meet your ends.
A sub-set I like to consider is:
-Consistency - a pattern of behavior or performing in a similar way, or always happening in a similar way.

E. Kant famously wrote in his Theoretical Epistemology.
“Thoughts without content are empty, Intuitions without concepts are blind"

Somedays I prefer art over engineering because one can create using a ”Free Floating Rationale” in art without too many negative consequence.

 

Emmanuel Kant had some dubious personal habits apparently.

 

Hi,

Thanks all for your answers.

My personal experience (automotive and marine industries) aligns pretty much to what @Dolfiman is describing in his post.

There is a spiral, nonlinear component in each of the phases. Still, the phases can help you remember how far you are in the analysis/design, as well as the inputs (requirements) and the desired outputs (deliverables with a predefined level of detail) for that phase. Also, at the end of each phase, you go through a decisional gate where it is decided whether to move forward into the next phase (engaging more resources) or stop (for example, if the project is deemed not to be feasible technically, economically, in quality, and/or time).

Also, my experience with the design spirals usually found in the bibliography is that they are very academic but can barely be put to work in the context of a real project, as if the first thing you have to do when you start a new project is to sit down to design a hull. They typically do not make evident the first steps of a project, which are more related to architectural analysis and prototyping: evaluation of areas, volumes, their connection, and flow based on the initial requirements.

The best design spiral I have found, from my point of view, for sure, which actually works in the scope of an actual project, is the one presented in the book "Architecture Navale" (in French). This is a picture of it (sorry for the bad quality):

 

Pablo,
Great example,
The design spiral helps me to identification desirable or undesirable feedback loops in the design process if this is what I think this digram shows.
With each new or different design spiral we add information/knowledge to our designing experiences and it can becomes second nature.
I used the design spiral in designing and building of furniture and other project through the entire process. It would help me see things that I may have neglected that might come back to have negative feedback loop in the end.

 

A really nice graph. What I don't understand is how you can make a "displacement approximation", study the "hull characteristics", a "power estimate" or even a "scantling" calculation, when you still don't have the body lines. plan that, according to that beautiful graph, is obtained several steps later.

 

The design spiral was invented for for a neck-top computer.
Its seem that the people who think the design spiral is worthless are using expensive computer design programs and less staff to optimize profit. This is big money Top Down approach to design and their efforts to automat work flow could make a large portion of the process easily replaced by a chatbot and chatbot creator in the future for a profit.
"Ships are the nearest things to dreams that hand" and minds "have ever made."
Robert N. Rose's
So much for all that kind poetic nonsense.

 

The lines plan, is the last thing to establish.
It all depends how you go about your process and what information you have to hand and what you are aiming at.

If you are designing say a fishing boat of say 20m, and your client also wants a fishing boat of say 20m, with similar duty but very different function, it would make sense to use the existing information.
But rarely in design does one get the luxury of using and exact hull form for a new design.

Whereas, if a client wanted say a 25m fishing boat and with more beam or speed etc, then the GA would be drawn up to establish the wants/needs of the client that forms the "design".
Then one would use the basis 20m hull's DNA information, such as weights, in whatever form one wishes, to begin the weight estimate tailored to the 25m.
All easily done without any Lines Plan, nor tank test results etc etc.

Naval architecture is is all about trends, not absolutes.

 

"Facts, facts", as one of my best friends says, not trends.

The last thing of which because, for example, I do not know how to determine, let's say, the distance between supports of a frame, in order to calculate its scantlings, without knowing the shapes of the boat.
I don't know how many fishing boats you have designed, I'm afraid none, but the reality is that two boats, made in GRP with the same mold, for different owners are never the same and sometimes not even similar, in many of their aspects.
But I will not discuss anything else with you who know perfectly well what I am talking about and you know perfectly well that I am right. Thank you, once again, for your "unforgettable" lessons.

 

All this shows is that you are an "absolute" person and are unable to join the dots - the trends - to get to the bigger picture.
That's fine, if that is the way you approach "design".

Again as an example, a fishing boat of say 20m, and then i need one of 25m.
It takes very little to simply divide the structure weight of the 20m into - per metre length - and then pro-rata to 25m.
This is called a trend. It is not an absolute, but a trend.
And within a very short period of time, one can establish a basic set of weights.
Simple.

Then it is not the same....Simple.

If you wish to over think everything to prove a point that does not exist.....go for it.
You're in an echo chamber.

 

The process is carried out more than once but not all of the operations need to be carried out each time you go through the spiral. You would update each segment more than once improving the quality of the information.
I use a spreadsheet to calculate the quantity of the attribute.
It is a teaching tool as well as a organization and management tool in "The Principles of Yacht Design"
http://protei.org/download/20110417Principles of yacht design - Larsson, Eliasson.pdf